1 | Some Arm bugfixes for rc2... | 1 | Handful of bugfixes for rc2. None of these are particularly critical |
---|---|---|---|
2 | or exciting. | ||
2 | 3 | ||
3 | thanks | ||
4 | -- PMM | 4 | -- PMM |
5 | 5 | ||
6 | The following changes since commit e6ebbd46b6e539f3613136111977721d212c2812: | 6 | The following changes since commit 45a150aa2b3492acf6691c7bdbeb25a8545d8345: |
7 | 7 | ||
8 | Merge remote-tracking branch 'remotes/kevin/tags/for-upstream' into staging (2018-11-19 14:31:48 +0000) | 8 | Merge remote-tracking branch 'remotes/ericb/tags/pull-bitmaps-2020-08-03' into staging (2020-08-03 15:13:49 +0100) |
9 | 9 | ||
10 | are available in the Git repository at: | 10 | are available in the Git repository at: |
11 | 11 | ||
12 | https://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20181119 | 12 | https://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20200803 |
13 | 13 | ||
14 | for you to fetch changes up to a00d7f2048c2a1a6a4487ac195c804c78adcf60e: | 14 | for you to fetch changes up to 13557fd392890cbd985bceba7f717e01efd674b8: |
15 | 15 | ||
16 | MAINTAINERS: list myself as maintainer for various Arm boards (2018-11-19 15:55:11 +0000) | 16 | hw/timer/imx_epit: Avoid assertion when CR.SWR is written (2020-08-03 17:56:11 +0100) |
17 | 17 | ||
18 | ---------------------------------------------------------------- | 18 | ---------------------------------------------------------------- |
19 | target-arm queue: | 19 | target-arm queue: |
20 | * various MAINTAINERS file updates | 20 | * hw/timer/imx_epit: Avoid assertion when CR.SWR is written |
21 | * hw/block/onenand: use qemu_log_mask() for reporting | 21 | * netduino2, netduinoplus2, microbit: set system_clock_scale so that |
22 | * hw/block/onenand: Fix off-by-one error allowing out-of-bounds read | 22 | SysTick running on the CPU clock works |
23 | on the n800 and n810 machine models | 23 | * target/arm: Avoid maybe-uninitialized warning with gcc 4.9 |
24 | * target/arm: fix smc incorrectly trapping to EL3 when secure is off | 24 | * target/arm: Fix AddPAC error indication |
25 | * hw/arm/stm32f205: Fix the UART and Timer region size | 25 | * Make AIRCR.SYSRESETREQ actually reset the system for the |
26 | * target/arm: read ID registers for KVM guests so they can be | 26 | microbit, mps2-*, musca-*, netduino* boards |
27 | used to gate "is feature X present" checks | ||
28 | 27 | ||
29 | ---------------------------------------------------------------- | 28 | ---------------------------------------------------------------- |
30 | Luc Michel (1): | 29 | Kaige Li (1): |
31 | target/arm: fix smc incorrectly trapping to EL3 when secure is off | 30 | target/arm: Avoid maybe-uninitialized warning with gcc 4.9 |
32 | 31 | ||
33 | Peter Maydell (3): | 32 | Peter Maydell (6): |
34 | hw/block/onenand: Fix off-by-one error allowing out-of-bounds read | 33 | hw/arm/netduino2, netduinoplus2: Set system_clock_scale |
35 | hw/block/onenand: use qemu_log_mask() for reporting | 34 | include/hw/irq.h: New function qemu_irq_is_connected() |
36 | MAINTAINERS: list myself as maintainer for various Arm boards | 35 | hw/intc/armv7m_nvic: Provide default "reset the system" behaviour for SYSRESETREQ |
36 | msf2-soc, stellaris: Don't wire up SYSRESETREQ | ||
37 | hw/arm/nrf51_soc: Set system_clock_scale | ||
38 | hw/timer/imx_epit: Avoid assertion when CR.SWR is written | ||
37 | 39 | ||
38 | Richard Henderson (4): | 40 | Richard Henderson (1): |
39 | target/arm: Install ARMISARegisters from kvm host | 41 | target/arm: Fix AddPAC error indication |
40 | target/arm: Fill in ARMISARegisters for kvm64 | ||
41 | target/arm: Introduce read_sys_reg32 for kvm32 | ||
42 | target/arm: Fill in ARMISARegisters for kvm32 | ||
43 | 42 | ||
44 | Seth Kintigh (1): | 43 | include/hw/arm/armv7m.h | 4 +++- |
45 | hw/arm/stm32f205: Fix the UART and Timer region size | 44 | include/hw/irq.h | 18 ++++++++++++++++++ |
45 | hw/arm/msf2-soc.c | 11 ----------- | ||
46 | hw/arm/netduino2.c | 10 ++++++++++ | ||
47 | hw/arm/netduinoplus2.c | 10 ++++++++++ | ||
48 | hw/arm/nrf51_soc.c | 5 +++++ | ||
49 | hw/arm/stellaris.c | 12 ------------ | ||
50 | hw/intc/armv7m_nvic.c | 17 ++++++++++++++++- | ||
51 | hw/timer/imx_epit.c | 13 ++++++++++--- | ||
52 | target/arm/pauth_helper.c | 6 +++++- | ||
53 | target/arm/translate-a64.c | 2 +- | ||
54 | tests/tcg/aarch64/pauth-5.c | 33 +++++++++++++++++++++++++++++++++ | ||
55 | tests/tcg/aarch64/Makefile.target | 2 +- | ||
56 | 13 files changed, 112 insertions(+), 31 deletions(-) | ||
57 | create mode 100644 tests/tcg/aarch64/pauth-5.c | ||
46 | 58 | ||
47 | Thomas Huth (1): | ||
48 | MAINTAINERS: Add entries for missing ARM boards | ||
49 | |||
50 | target/arm/kvm_arm.h | 1 + | ||
51 | hw/block/onenand.c | 24 +++++----- | ||
52 | hw/char/stm32f2xx_usart.c | 2 +- | ||
53 | hw/timer/stm32f2xx_timer.c | 2 +- | ||
54 | target/arm/kvm.c | 1 + | ||
55 | target/arm/kvm32.c | 77 ++++++++++++++++++++------------ | ||
56 | target/arm/kvm64.c | 90 +++++++++++++++++++++++++++++++++++++- | ||
57 | target/arm/op_helper.c | 54 +++++++++++++++++++---- | ||
58 | MAINTAINERS | 106 +++++++++++++++++++++++++++++++++++++++------ | ||
59 | 9 files changed, 293 insertions(+), 64 deletions(-) | ||
60 | diff view generated by jsdifflib |
1 | From: Luc Michel <luc.michel@greensocs.com> | 1 | The netduino2 and netduinoplus2 boards forgot to set the system_clock_scale |
---|---|---|---|
2 | global, which meant that if guest code used the systick timer in "use | ||
3 | the processor clock" mode it would hang because time never advances. | ||
2 | 4 | ||
3 | This commit fixes a case where the CPU would try to go to EL3 when | 5 | Set the global to match the documented CPU clock speed of these boards. |
4 | executing an smc instruction, even though ARM_FEATURE_EL3 is false. This | 6 | Judging by the data sheet this is slightly simplistic because the |
5 | case is raised when the PSCI conduit is set to smc, but the smc | 7 | SoC allows configuration of the SYSCLK source and frequency via the |
6 | instruction does not lead to a valid PSCI call. | 8 | RCC (reset and clock control) module, but we don't model that. |
7 | 9 | ||
8 | QEMU crashes with an assertion failure latter on because of incoherent | 10 | Fixes: https://bugs.launchpad.net/qemu/+bug/1876187 |
9 | mmu_idx. | 11 | Signed-off-by: Peter Maydell <peter.maydell@linaro.org> |
12 | Reviewed-by: Alistair Francis <alistair.francis@wdc.com> | ||
13 | Message-id: 20200727162617.26227-1-peter.maydell@linaro.org | ||
14 | --- | ||
15 | hw/arm/netduino2.c | 10 ++++++++++ | ||
16 | hw/arm/netduinoplus2.c | 10 ++++++++++ | ||
17 | 2 files changed, 20 insertions(+) | ||
10 | 18 | ||
11 | This commit refactors the pre_smc helper by enumerating all the possible | 19 | diff --git a/hw/arm/netduino2.c b/hw/arm/netduino2.c |
12 | way of handling an scm instruction, and covering the previously missing | ||
13 | case leading to the crash. | ||
14 | |||
15 | The following minimal test would crash before this commit: | ||
16 | |||
17 | .global _start | ||
18 | .text | ||
19 | _start: | ||
20 | ldr x0, =0xdeadbeef ; invalid PSCI call | ||
21 | smc #0 | ||
22 | |||
23 | run with the following command line: | ||
24 | |||
25 | aarch64-linux-gnu-gcc -nostdinc -nostdlib -Wl,-Ttext=40000000 \ | ||
26 | -o test test.s | ||
27 | |||
28 | qemu-system-aarch64 -M virt,virtualization=on,secure=off \ | ||
29 | -cpu cortex-a57 -kernel test | ||
30 | |||
31 | Signed-off-by: Luc Michel <luc.michel@greensocs.com> | ||
32 | Message-id: 20181117160213.18995-1-luc.michel@greensocs.com | ||
33 | Reviewed-by: Peter Maydell <peter.maydell@linaro.org> | ||
34 | Signed-off-by: Peter Maydell <peter.maydell@linaro.org> | ||
35 | --- | ||
36 | target/arm/op_helper.c | 54 +++++++++++++++++++++++++++++++++++------- | ||
37 | 1 file changed, 46 insertions(+), 8 deletions(-) | ||
38 | |||
39 | diff --git a/target/arm/op_helper.c b/target/arm/op_helper.c | ||
40 | index XXXXXXX..XXXXXXX 100644 | 20 | index XXXXXXX..XXXXXXX 100644 |
41 | --- a/target/arm/op_helper.c | 21 | --- a/hw/arm/netduino2.c |
42 | +++ b/target/arm/op_helper.c | 22 | +++ b/hw/arm/netduino2.c |
43 | @@ -XXX,XX +XXX,XX @@ void HELPER(pre_smc)(CPUARMState *env, uint32_t syndrome) | 23 | @@ -XXX,XX +XXX,XX @@ |
44 | ARMCPU *cpu = arm_env_get_cpu(env); | 24 | #include "hw/arm/stm32f205_soc.h" |
45 | int cur_el = arm_current_el(env); | 25 | #include "hw/arm/boot.h" |
46 | bool secure = arm_is_secure(env); | 26 | |
47 | - bool smd = env->cp15.scr_el3 & SCR_SMD; | 27 | +/* Main SYSCLK frequency in Hz (120MHz) */ |
48 | + bool smd_flag = env->cp15.scr_el3 & SCR_SMD; | 28 | +#define SYSCLK_FRQ 120000000ULL |
49 | + | 29 | + |
30 | static void netduino2_init(MachineState *machine) | ||
31 | { | ||
32 | DeviceState *dev; | ||
33 | |||
50 | + /* | 34 | + /* |
51 | + * SMC behaviour is summarized in the following table. | 35 | + * TODO: ideally we would model the SoC RCC and let it handle |
52 | + * This helper handles the "Trap to EL2" and "Undef insn" cases. | 36 | + * system_clock_scale, including its ability to define different |
53 | + * The "Trap to EL3" and "PSCI call" cases are handled in the exception | 37 | + * possible SYSCLK sources. |
54 | + * helper. | ||
55 | + * | ||
56 | + * -> ARM_FEATURE_EL3 and !SMD | ||
57 | + * HCR_TSC && NS EL1 !HCR_TSC || !NS EL1 | ||
58 | + * | ||
59 | + * Conduit SMC, valid call Trap to EL2 PSCI Call | ||
60 | + * Conduit SMC, inval call Trap to EL2 Trap to EL3 | ||
61 | + * Conduit not SMC Trap to EL2 Trap to EL3 | ||
62 | + * | ||
63 | + * | ||
64 | + * -> ARM_FEATURE_EL3 and SMD | ||
65 | + * HCR_TSC && NS EL1 !HCR_TSC || !NS EL1 | ||
66 | + * | ||
67 | + * Conduit SMC, valid call Trap to EL2 PSCI Call | ||
68 | + * Conduit SMC, inval call Trap to EL2 Undef insn | ||
69 | + * Conduit not SMC Trap to EL2 Undef insn | ||
70 | + * | ||
71 | + * | ||
72 | + * -> !ARM_FEATURE_EL3 | ||
73 | + * HCR_TSC && NS EL1 !HCR_TSC || !NS EL1 | ||
74 | + * | ||
75 | + * Conduit SMC, valid call Trap to EL2 PSCI Call | ||
76 | + * Conduit SMC, inval call Trap to EL2 Undef insn | ||
77 | + * Conduit not SMC Undef insn Undef insn | ||
78 | + */ | 38 | + */ |
39 | + system_clock_scale = NANOSECONDS_PER_SECOND / SYSCLK_FRQ; | ||
79 | + | 40 | + |
80 | /* On ARMv8 with EL3 AArch64, SMD applies to both S and NS state. | 41 | dev = qdev_new(TYPE_STM32F205_SOC); |
81 | * On ARMv8 with EL3 AArch32, or ARMv7 with the Virtualization | 42 | qdev_prop_set_string(dev, "cpu-type", ARM_CPU_TYPE_NAME("cortex-m3")); |
82 | * extensions, SMD only applies to NS state. | 43 | sysbus_realize_and_unref(SYS_BUS_DEVICE(dev), &error_fatal); |
83 | @@ -XXX,XX +XXX,XX @@ void HELPER(pre_smc)(CPUARMState *env, uint32_t syndrome) | 44 | diff --git a/hw/arm/netduinoplus2.c b/hw/arm/netduinoplus2.c |
84 | * doesn't exist, but we forbid the guest to set it to 1 in scr_write(), | 45 | index XXXXXXX..XXXXXXX 100644 |
85 | * so we need not special case this here. | 46 | --- a/hw/arm/netduinoplus2.c |
86 | */ | 47 | +++ b/hw/arm/netduinoplus2.c |
87 | - bool undef = arm_feature(env, ARM_FEATURE_AARCH64) ? smd : smd && !secure; | 48 | @@ -XXX,XX +XXX,XX @@ |
88 | + bool smd = arm_feature(env, ARM_FEATURE_AARCH64) ? smd_flag | 49 | #include "hw/arm/stm32f405_soc.h" |
89 | + : smd_flag && !secure; | 50 | #include "hw/arm/boot.h" |
90 | 51 | ||
91 | if (!arm_feature(env, ARM_FEATURE_EL3) && | 52 | +/* Main SYSCLK frequency in Hz (168MHz) */ |
92 | cpu->psci_conduit != QEMU_PSCI_CONDUIT_SMC) { | 53 | +#define SYSCLK_FRQ 168000000ULL |
93 | @@ -XXX,XX +XXX,XX @@ void HELPER(pre_smc)(CPUARMState *env, uint32_t syndrome) | ||
94 | * to forbid its EL1 from making PSCI calls into QEMU's | ||
95 | * "firmware" via HCR.TSC, so for these purposes treat | ||
96 | * PSCI-via-SMC as implying an EL3. | ||
97 | + * This handles the very last line of the previous table. | ||
98 | */ | ||
99 | - undef = true; | ||
100 | - } else if (!secure && cur_el == 1 && (env->cp15.hcr_el2 & HCR_TSC)) { | ||
101 | + raise_exception(env, EXCP_UDEF, syn_uncategorized(), | ||
102 | + exception_target_el(env)); | ||
103 | + } | ||
104 | + | 54 | + |
105 | + if (!secure && cur_el == 1 && (env->cp15.hcr_el2 & HCR_TSC)) { | 55 | static void netduinoplus2_init(MachineState *machine) |
106 | /* In NS EL1, HCR controlled routing to EL2 has priority over SMD. | 56 | { |
107 | * We also want an EL2 guest to be able to forbid its EL1 from | 57 | DeviceState *dev; |
108 | * making PSCI calls into QEMU's "firmware" via HCR.TSC. | 58 | |
109 | + * This handles all the "Trap to EL2" cases of the previous table. | 59 | + /* |
110 | */ | 60 | + * TODO: ideally we would model the SoC RCC and let it handle |
111 | raise_exception(env, EXCP_HYP_TRAP, syndrome, 2); | 61 | + * system_clock_scale, including its ability to define different |
112 | } | 62 | + * possible SYSCLK sources. |
113 | 63 | + */ | |
114 | - /* If PSCI is enabled and this looks like a valid PSCI call then | 64 | + system_clock_scale = NANOSECONDS_PER_SECOND / SYSCLK_FRQ; |
115 | - * suppress the UNDEF -- we'll catch the SMC exception and | 65 | + |
116 | - * implement the PSCI call behaviour there. | 66 | dev = qdev_new(TYPE_STM32F405_SOC); |
117 | + /* Catch the two remaining "Undef insn" cases of the previous table: | 67 | qdev_prop_set_string(dev, "cpu-type", ARM_CPU_TYPE_NAME("cortex-m4")); |
118 | + * - PSCI conduit is SMC but we don't have a valid PCSI call, | 68 | sysbus_realize_and_unref(SYS_BUS_DEVICE(dev), &error_fatal); |
119 | + * - We don't have EL3 or SMD is set. | ||
120 | */ | ||
121 | - if (undef && !arm_is_psci_call(cpu, EXCP_SMC)) { | ||
122 | + if (!arm_is_psci_call(cpu, EXCP_SMC) && | ||
123 | + (smd || !arm_feature(env, ARM_FEATURE_EL3))) { | ||
124 | raise_exception(env, EXCP_UDEF, syn_uncategorized(), | ||
125 | exception_target_el(env)); | ||
126 | } | ||
127 | -- | 69 | -- |
128 | 2.19.1 | 70 | 2.20.1 |
129 | 71 | ||
130 | 72 | diff view generated by jsdifflib |
1 | From: Seth Kintigh <skintigh@gmail.com> | 1 | Mostly devices don't need to care whether one of their output |
---|---|---|---|
2 | qemu_irq lines is connected, because functions like qemu_set_irq() | ||
3 | silently do nothing if there is nothing on the other end. However | ||
4 | sometimes a device might want to implement default behaviour for the | ||
5 | case where the machine hasn't wired the line up to anywhere. | ||
2 | 6 | ||
3 | The UART and timer devices for the stm32f205 were being created | 7 | Provide a function qemu_irq_is_connected() that devices can use for |
4 | with memory regions that were too large. Use the size specified | 8 | this purpose. (The test is trivial but encapsulating it in a |
5 | in the chip datasheet. | 9 | function makes it easier to see where we're doing it in case we need |
10 | to change the implementation later.) | ||
6 | 11 | ||
7 | The old sizes were so large that the devices would overlap with | 12 | Signed-off-by: Peter Maydell <peter.maydell@linaro.org> |
8 | each other in the SoC memory map, so this fixes a bug that | 13 | Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> |
9 | caused odd behavior and/or crashes when trying to set up multiple | 14 | Reviewed-by: Alistair Francis <alistair.francis@wdc.com> |
10 | UARTs. | 15 | Message-id: 20200728103744.6909-2-peter.maydell@linaro.org |
16 | --- | ||
17 | include/hw/irq.h | 18 ++++++++++++++++++ | ||
18 | 1 file changed, 18 insertions(+) | ||
11 | 19 | ||
12 | Signed-off-by: Seth Kintigh <skintigh@gmail.com> | 20 | diff --git a/include/hw/irq.h b/include/hw/irq.h |
13 | Reviewed-by: Peter Maydell <peter.maydell@linaro.org> | ||
14 | [PMM: rephrased commit message to follow our usual standard] | ||
15 | Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> | ||
16 | Tested-by: Philippe Mathieu-Daudé <f4bug@amsat.org> | ||
17 | Signed-off-by: Peter Maydell <peter.maydell@linaro.org> | ||
18 | --- | ||
19 | hw/char/stm32f2xx_usart.c | 2 +- | ||
20 | hw/timer/stm32f2xx_timer.c | 2 +- | ||
21 | 2 files changed, 2 insertions(+), 2 deletions(-) | ||
22 | |||
23 | diff --git a/hw/char/stm32f2xx_usart.c b/hw/char/stm32f2xx_usart.c | ||
24 | index XXXXXXX..XXXXXXX 100644 | 21 | index XXXXXXX..XXXXXXX 100644 |
25 | --- a/hw/char/stm32f2xx_usart.c | 22 | --- a/include/hw/irq.h |
26 | +++ b/hw/char/stm32f2xx_usart.c | 23 | +++ b/include/hw/irq.h |
27 | @@ -XXX,XX +XXX,XX @@ static void stm32f2xx_usart_init(Object *obj) | 24 | @@ -XXX,XX +XXX,XX @@ qemu_irq qemu_irq_split(qemu_irq irq1, qemu_irq irq2); |
28 | sysbus_init_irq(SYS_BUS_DEVICE(obj), &s->irq); | 25 | on an existing vector of qemu_irq. */ |
29 | 26 | void qemu_irq_intercept_in(qemu_irq *gpio_in, qemu_irq_handler handler, int n); | |
30 | memory_region_init_io(&s->mmio, obj, &stm32f2xx_usart_ops, s, | 27 | |
31 | - TYPE_STM32F2XX_USART, 0x2000); | 28 | +/** |
32 | + TYPE_STM32F2XX_USART, 0x400); | 29 | + * qemu_irq_is_connected: Return true if IRQ line is wired up |
33 | sysbus_init_mmio(SYS_BUS_DEVICE(obj), &s->mmio); | 30 | + * |
34 | } | 31 | + * If a qemu_irq has a device on the other (receiving) end of it, |
35 | 32 | + * return true; otherwise return false. | |
36 | diff --git a/hw/timer/stm32f2xx_timer.c b/hw/timer/stm32f2xx_timer.c | 33 | + * |
37 | index XXXXXXX..XXXXXXX 100644 | 34 | + * Usually device models don't need to care whether the machine model |
38 | --- a/hw/timer/stm32f2xx_timer.c | 35 | + * has wired up their outbound qemu_irq lines, because functions like |
39 | +++ b/hw/timer/stm32f2xx_timer.c | 36 | + * qemu_set_irq() silently do nothing if there is nothing on the other |
40 | @@ -XXX,XX +XXX,XX @@ static void stm32f2xx_timer_init(Object *obj) | 37 | + * end of the line. However occasionally a device model will want to |
41 | sysbus_init_irq(SYS_BUS_DEVICE(obj), &s->irq); | 38 | + * provide default behaviour if its output is left floating, and |
42 | 39 | + * it can use this function to identify when that is the case. | |
43 | memory_region_init_io(&s->iomem, obj, &stm32f2xx_timer_ops, s, | 40 | + */ |
44 | - "stm32f2xx_timer", 0x4000); | 41 | +static inline bool qemu_irq_is_connected(qemu_irq irq) |
45 | + "stm32f2xx_timer", 0x400); | 42 | +{ |
46 | sysbus_init_mmio(SYS_BUS_DEVICE(obj), &s->iomem); | 43 | + return irq != NULL; |
47 | 44 | +} | |
48 | s->timer = timer_new_ns(QEMU_CLOCK_VIRTUAL, stm32f2xx_timer_interrupt, s); | 45 | + |
46 | #endif | ||
49 | -- | 47 | -- |
50 | 2.19.1 | 48 | 2.20.1 |
51 | 49 | ||
52 | 50 | diff view generated by jsdifflib |
1 | From: Richard Henderson <richard.henderson@linaro.org> | 1 | The NVIC provides an outbound qemu_irq "SYSRESETREQ" which it signals |
---|---|---|---|
2 | when the guest sets the SYSRESETREQ bit in the AIRCR register. This | ||
3 | matches the hardware design (where the CPU has a signal of this name | ||
4 | and it is up to the SoC to connect that up to an actual reset | ||
5 | mechanism), but in QEMU it mostly results in duplicated code in SoC | ||
6 | objects and bugs where SoC model implementors forget to wire up the | ||
7 | SYSRESETREQ line. | ||
2 | 8 | ||
3 | Assert that the value to be written is the correct size. | 9 | Provide a default behaviour for the case where SYSRESETREQ is not |
4 | No change in functionality here, just mirroring the same | 10 | actually connected to anything: use qemu_system_reset_request() to |
5 | function from kvm64. | 11 | perform a system reset. This will allow us to remove the |
12 | implementations of SYSRESETREQ handling from the boards where that's | ||
13 | exactly what it does, and also fixes the bugs in the board models | ||
14 | which forgot to wire up the signal: | ||
6 | 15 | ||
7 | Signed-off-by: Richard Henderson <richard.henderson@linaro.org> | 16 | * microbit |
8 | Message-id: 20181113180154.17903-4-richard.henderson@linaro.org | 17 | * mps2-an385 |
9 | Reviewed-by: Peter Maydell <peter.maydell@linaro.org> | 18 | * mps2-an505 |
19 | * mps2-an511 | ||
20 | * mps2-an521 | ||
21 | * musca-a | ||
22 | * musca-b1 | ||
23 | * netduino | ||
24 | * netduinoplus2 | ||
25 | |||
26 | We still allow the board to wire up the signal if it needs to, in case | ||
27 | we need to model more complicated reset controller logic or to model | ||
28 | buggy SoC hardware which forgot to wire up the line itself. But | ||
29 | defaulting to "reset the system" is more often going to be correct | ||
30 | than defaulting to "do nothing". | ||
31 | |||
10 | Signed-off-by: Peter Maydell <peter.maydell@linaro.org> | 32 | Signed-off-by: Peter Maydell <peter.maydell@linaro.org> |
33 | Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> | ||
34 | Reviewed-by: Alistair Francis <alistair.francis@wdc.com> | ||
35 | Message-id: 20200728103744.6909-3-peter.maydell@linaro.org | ||
11 | --- | 36 | --- |
12 | target/arm/kvm32.c | 41 ++++++++++++++++------------------------- | 37 | include/hw/arm/armv7m.h | 4 +++- |
13 | 1 file changed, 16 insertions(+), 25 deletions(-) | 38 | hw/intc/armv7m_nvic.c | 17 ++++++++++++++++- |
39 | 2 files changed, 19 insertions(+), 2 deletions(-) | ||
14 | 40 | ||
15 | diff --git a/target/arm/kvm32.c b/target/arm/kvm32.c | 41 | diff --git a/include/hw/arm/armv7m.h b/include/hw/arm/armv7m.h |
16 | index XXXXXXX..XXXXXXX 100644 | 42 | index XXXXXXX..XXXXXXX 100644 |
17 | --- a/target/arm/kvm32.c | 43 | --- a/include/hw/arm/armv7m.h |
18 | +++ b/target/arm/kvm32.c | 44 | +++ b/include/hw/arm/armv7m.h |
19 | @@ -XXX,XX +XXX,XX @@ static inline void set_feature(uint64_t *features, int feature) | 45 | @@ -XXX,XX +XXX,XX @@ typedef struct { |
20 | *features |= 1ULL << feature; | 46 | |
21 | } | 47 | /* ARMv7M container object. |
22 | 48 | * + Unnamed GPIO input lines: external IRQ lines for the NVIC | |
23 | +static int read_sys_reg32(int fd, uint32_t *pret, uint64_t id) | 49 | - * + Named GPIO output SYSRESETREQ: signalled for guest AIRCR.SYSRESETREQ |
50 | + * + Named GPIO output SYSRESETREQ: signalled for guest AIRCR.SYSRESETREQ. | ||
51 | + * If this GPIO is not wired up then the NVIC will default to performing | ||
52 | + * a qemu_system_reset_request(SHUTDOWN_CAUSE_GUEST_RESET). | ||
53 | * + Property "cpu-type": CPU type to instantiate | ||
54 | * + Property "num-irq": number of external IRQ lines | ||
55 | * + Property "memory": MemoryRegion defining the physical address space | ||
56 | diff --git a/hw/intc/armv7m_nvic.c b/hw/intc/armv7m_nvic.c | ||
57 | index XXXXXXX..XXXXXXX 100644 | ||
58 | --- a/hw/intc/armv7m_nvic.c | ||
59 | +++ b/hw/intc/armv7m_nvic.c | ||
60 | @@ -XXX,XX +XXX,XX @@ | ||
61 | #include "hw/intc/armv7m_nvic.h" | ||
62 | #include "hw/irq.h" | ||
63 | #include "hw/qdev-properties.h" | ||
64 | +#include "sysemu/runstate.h" | ||
65 | #include "target/arm/cpu.h" | ||
66 | #include "exec/exec-all.h" | ||
67 | #include "exec/memop.h" | ||
68 | @@ -XXX,XX +XXX,XX @@ static const uint8_t nvic_id[] = { | ||
69 | 0x00, 0xb0, 0x1b, 0x00, 0x0d, 0xe0, 0x05, 0xb1 | ||
70 | }; | ||
71 | |||
72 | +static void signal_sysresetreq(NVICState *s) | ||
24 | +{ | 73 | +{ |
25 | + struct kvm_one_reg idreg = { .id = id, .addr = (uintptr_t)pret }; | 74 | + if (qemu_irq_is_connected(s->sysresetreq)) { |
26 | + | 75 | + qemu_irq_pulse(s->sysresetreq); |
27 | + assert((id & KVM_REG_SIZE_MASK) == KVM_REG_SIZE_U32); | 76 | + } else { |
28 | + return ioctl(fd, KVM_GET_ONE_REG, &idreg); | 77 | + /* |
78 | + * Default behaviour if the SoC doesn't need to wire up | ||
79 | + * SYSRESETREQ (eg to a system reset controller of some kind): | ||
80 | + * perform a system reset via the usual QEMU API. | ||
81 | + */ | ||
82 | + qemu_system_reset_request(SHUTDOWN_CAUSE_GUEST_RESET); | ||
83 | + } | ||
29 | +} | 84 | +} |
30 | + | 85 | + |
31 | bool kvm_arm_get_host_cpu_features(ARMHostCPUFeatures *ahcf) | 86 | static int nvic_pending_prio(NVICState *s) |
32 | { | 87 | { |
33 | /* Identify the feature bits corresponding to the host CPU, and | 88 | /* return the group priority of the current pending interrupt, |
34 | @@ -XXX,XX +XXX,XX @@ bool kvm_arm_get_host_cpu_features(ARMHostCPUFeatures *ahcf) | 89 | @@ -XXX,XX +XXX,XX @@ static void nvic_writel(NVICState *s, uint32_t offset, uint32_t value, |
35 | * we have to create a scratch VM, create a single CPU inside it, | 90 | if (value & R_V7M_AIRCR_SYSRESETREQ_MASK) { |
36 | * and then query that CPU for the relevant ID registers. | 91 | if (attrs.secure || |
37 | */ | 92 | !(cpu->env.v7m.aircr & R_V7M_AIRCR_SYSRESETREQS_MASK)) { |
38 | - int i, ret, fdarray[3]; | 93 | - qemu_irq_pulse(s->sysresetreq); |
39 | + int err = 0, fdarray[3]; | 94 | + signal_sysresetreq(s); |
40 | uint32_t midr, id_pfr0, mvfr1; | 95 | } |
41 | uint64_t features = 0; | 96 | } |
42 | + | 97 | if (value & R_V7M_AIRCR_VECTCLRACTIVE_MASK) { |
43 | /* Old kernels may not know about the PREFERRED_TARGET ioctl: however | ||
44 | * we know these will only support creating one kind of guest CPU, | ||
45 | * which is its preferred CPU type. | ||
46 | @@ -XXX,XX +XXX,XX @@ bool kvm_arm_get_host_cpu_features(ARMHostCPUFeatures *ahcf) | ||
47 | QEMU_KVM_ARM_TARGET_NONE | ||
48 | }; | ||
49 | struct kvm_vcpu_init init; | ||
50 | - struct kvm_one_reg idregs[] = { | ||
51 | - { | ||
52 | - .id = KVM_REG_ARM | KVM_REG_SIZE_U32 | ||
53 | - | ENCODE_CP_REG(15, 0, 0, 0, 0, 0, 0), | ||
54 | - .addr = (uintptr_t)&midr, | ||
55 | - }, | ||
56 | - { | ||
57 | - .id = KVM_REG_ARM | KVM_REG_SIZE_U32 | ||
58 | - | ENCODE_CP_REG(15, 0, 0, 0, 1, 0, 0), | ||
59 | - .addr = (uintptr_t)&id_pfr0, | ||
60 | - }, | ||
61 | - { | ||
62 | - .id = KVM_REG_ARM | KVM_REG_SIZE_U32 | ||
63 | - | KVM_REG_ARM_VFP | KVM_REG_ARM_VFP_MVFR1, | ||
64 | - .addr = (uintptr_t)&mvfr1, | ||
65 | - }, | ||
66 | - }; | ||
67 | |||
68 | if (!kvm_arm_create_scratch_host_vcpu(cpus_to_try, fdarray, &init)) { | ||
69 | return false; | ||
70 | @@ -XXX,XX +XXX,XX @@ bool kvm_arm_get_host_cpu_features(ARMHostCPUFeatures *ahcf) | ||
71 | */ | ||
72 | ahcf->dtb_compatible = "arm,arm-v7"; | ||
73 | |||
74 | - for (i = 0; i < ARRAY_SIZE(idregs); i++) { | ||
75 | - ret = ioctl(fdarray[2], KVM_GET_ONE_REG, &idregs[i]); | ||
76 | - if (ret) { | ||
77 | - break; | ||
78 | - } | ||
79 | - } | ||
80 | + err |= read_sys_reg32(fdarray[2], &midr, ARM_CP15_REG32(0, 0, 0, 0)); | ||
81 | + err |= read_sys_reg32(fdarray[2], &id_pfr0, ARM_CP15_REG32(0, 0, 1, 0)); | ||
82 | + err |= read_sys_reg32(fdarray[2], &mvfr1, | ||
83 | + KVM_REG_ARM | KVM_REG_SIZE_U32 | | ||
84 | + KVM_REG_ARM_VFP | KVM_REG_ARM_VFP_MVFR1); | ||
85 | |||
86 | kvm_arm_destroy_scratch_host_vcpu(fdarray); | ||
87 | |||
88 | - if (ret) { | ||
89 | + if (err < 0) { | ||
90 | return false; | ||
91 | } | ||
92 | |||
93 | -- | 98 | -- |
94 | 2.19.1 | 99 | 2.20.1 |
95 | 100 | ||
96 | 101 | diff view generated by jsdifflib |
1 | Update the onenand device to use qemu_log_mask() for reporting | 1 | The MSF2 SoC model and the Stellaris board code both wire |
---|---|---|---|
2 | guest errors and unimplemented features, rather than plain | 2 | SYSRESETREQ up to a function that just invokes |
3 | fprintf() and hw_error(). | 3 | qemu_system_reset_request(SHUTDOWN_CAUSE_GUEST_RESET); |
4 | 4 | This is now the default action that the NVIC does if the line is | |
5 | (We leave the hw_error() in onenand_reset(), as that is | 5 | not connected, so we can delete the handling code. |
6 | triggered by a failure to read the underlying block device | ||
7 | for the bootRAM, not by guest action.) | ||
8 | 6 | ||
9 | Signed-off-by: Peter Maydell <peter.maydell@linaro.org> | 7 | Signed-off-by: Peter Maydell <peter.maydell@linaro.org> |
10 | Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> | 8 | Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> |
11 | Reviewed-by: Richard Henderson <richard.henderson@linaro.org> | 9 | Reviewed-by: Alistair Francis <alistair.francis@wdc.com> |
12 | Reviewed-by: Thomas Huth <thuth@redhat.com> | 10 | Message-id: 20200728103744.6909-4-peter.maydell@linaro.org |
13 | Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com> | ||
14 | Message-id: 20181115143535.5885-3-peter.maydell@linaro.org | ||
15 | --- | 11 | --- |
16 | hw/block/onenand.c | 22 +++++++++++++--------- | 12 | hw/arm/msf2-soc.c | 11 ----------- |
17 | 1 file changed, 13 insertions(+), 9 deletions(-) | 13 | hw/arm/stellaris.c | 12 ------------ |
14 | 2 files changed, 23 deletions(-) | ||
18 | 15 | ||
19 | diff --git a/hw/block/onenand.c b/hw/block/onenand.c | 16 | diff --git a/hw/arm/msf2-soc.c b/hw/arm/msf2-soc.c |
20 | index XXXXXXX..XXXXXXX 100644 | 17 | index XXXXXXX..XXXXXXX 100644 |
21 | --- a/hw/block/onenand.c | 18 | --- a/hw/arm/msf2-soc.c |
22 | +++ b/hw/block/onenand.c | 19 | +++ b/hw/arm/msf2-soc.c |
23 | @@ -XXX,XX +XXX,XX @@ | 20 | @@ -XXX,XX +XXX,XX @@ |
24 | #include "exec/memory.h" | 21 | #include "hw/irq.h" |
25 | #include "hw/sysbus.h" | 22 | #include "hw/arm/msf2-soc.h" |
26 | #include "qemu/error-report.h" | 23 | #include "hw/misc/unimp.h" |
27 | +#include "qemu/log.h" | 24 | -#include "sysemu/runstate.h" |
28 | 25 | #include "sysemu/sysemu.h" | |
29 | /* 11 for 2kB-page OneNAND ("2nd generation") and 10 for 1kB-page chips */ | 26 | |
30 | #define PAGE_SHIFT 11 | 27 | #define MSF2_TIMER_BASE 0x40004000 |
31 | @@ -XXX,XX +XXX,XX @@ static void onenand_command(OneNANDState *s) | 28 | @@ -XXX,XX +XXX,XX @@ static const int spi_irq[MSF2_NUM_SPIS] = { 2, 3 }; |
32 | default: | 29 | static const int uart_irq[MSF2_NUM_UARTS] = { 10, 11 }; |
33 | s->status |= ONEN_ERR_CMD; | 30 | static const int timer_irq[MSF2_NUM_TIMERS] = { 14, 15 }; |
34 | s->intstatus |= ONEN_INT; | 31 | |
35 | - fprintf(stderr, "%s: unknown OneNAND command %x\n", | 32 | -static void do_sys_reset(void *opaque, int n, int level) |
36 | - __func__, s->command); | 33 | -{ |
37 | + qemu_log_mask(LOG_GUEST_ERROR, "unknown OneNAND command %x\n", | 34 | - if (level) { |
38 | + s->command); | 35 | - qemu_system_reset_request(SHUTDOWN_CAUSE_GUEST_RESET); |
36 | - } | ||
37 | -} | ||
38 | - | ||
39 | static void m2sxxx_soc_initfn(Object *obj) | ||
40 | { | ||
41 | MSF2State *s = MSF2_SOC(obj); | ||
42 | @@ -XXX,XX +XXX,XX @@ static void m2sxxx_soc_realize(DeviceState *dev_soc, Error **errp) | ||
43 | return; | ||
39 | } | 44 | } |
40 | 45 | ||
41 | onenand_intr_update(s); | 46 | - qdev_connect_gpio_out_named(DEVICE(&s->armv7m.nvic), "SYSRESETREQ", 0, |
42 | @@ -XXX,XX +XXX,XX @@ static uint64_t onenand_read(void *opaque, hwaddr addr, | 47 | - qemu_allocate_irq(&do_sys_reset, NULL, 0)); |
43 | case 0xff02: /* ECC Result of spare area data */ | 48 | - |
44 | case 0xff03: /* ECC Result of main area data */ | 49 | system_clock_scale = NANOSECONDS_PER_SECOND / s->m3clk; |
45 | case 0xff04: /* ECC Result of spare area data */ | 50 | |
46 | - hw_error("%s: implement ECC\n", __func__); | 51 | for (i = 0; i < MSF2_NUM_UARTS; i++) { |
47 | + qemu_log_mask(LOG_UNIMP, | 52 | diff --git a/hw/arm/stellaris.c b/hw/arm/stellaris.c |
48 | + "onenand: ECC result registers unimplemented\n"); | 53 | index XXXXXXX..XXXXXXX 100644 |
49 | return 0x0000; | 54 | --- a/hw/arm/stellaris.c |
50 | } | 55 | +++ b/hw/arm/stellaris.c |
51 | 56 | @@ -XXX,XX +XXX,XX @@ | |
52 | - fprintf(stderr, "%s: unknown OneNAND register %x\n", | 57 | #include "hw/boards.h" |
53 | - __func__, offset); | 58 | #include "qemu/log.h" |
54 | + qemu_log_mask(LOG_GUEST_ERROR, "read of unknown OneNAND register 0x%x\n", | 59 | #include "exec/address-spaces.h" |
55 | + offset); | 60 | -#include "sysemu/runstate.h" |
56 | return 0; | 61 | #include "sysemu/sysemu.h" |
62 | #include "hw/arm/armv7m.h" | ||
63 | #include "hw/char/pl011.h" | ||
64 | @@ -XXX,XX +XXX,XX @@ static void stellaris_adc_init(Object *obj) | ||
65 | qdev_init_gpio_in(dev, stellaris_adc_trigger, 1); | ||
57 | } | 66 | } |
58 | 67 | ||
59 | @@ -XXX,XX +XXX,XX @@ static void onenand_write(void *opaque, hwaddr addr, | 68 | -static |
60 | break; | 69 | -void do_sys_reset(void *opaque, int n, int level) |
61 | 70 | -{ | |
62 | default: | 71 | - if (level) { |
63 | - fprintf(stderr, "%s: unknown OneNAND boot command %"PRIx64"\n", | 72 | - qemu_system_reset_request(SHUTDOWN_CAUSE_GUEST_RESET); |
64 | - __func__, value); | 73 | - } |
65 | + qemu_log_mask(LOG_GUEST_ERROR, | 74 | -} |
66 | + "unknown OneNAND boot command %" PRIx64 "\n", | 75 | - |
67 | + value); | 76 | /* Board init. */ |
68 | } | 77 | static stellaris_board_info stellaris_boards[] = { |
69 | break; | 78 | { "LM3S811EVB", |
70 | 79 | @@ -XXX,XX +XXX,XX @@ static void stellaris_init(MachineState *ms, stellaris_board_info *board) | |
71 | @@ -XXX,XX +XXX,XX @@ static void onenand_write(void *opaque, hwaddr addr, | 80 | /* This will exit with an error if the user passed us a bad cpu_type */ |
72 | break; | 81 | sysbus_realize_and_unref(SYS_BUS_DEVICE(nvic), &error_fatal); |
73 | 82 | ||
74 | default: | 83 | - qdev_connect_gpio_out_named(nvic, "SYSRESETREQ", 0, |
75 | - fprintf(stderr, "%s: unknown OneNAND register %x\n", | 84 | - qemu_allocate_irq(&do_sys_reset, NULL, 0)); |
76 | - __func__, offset); | 85 | - |
77 | + qemu_log_mask(LOG_GUEST_ERROR, | 86 | if (board->dc1 & (1 << 16)) { |
78 | + "write to unknown OneNAND register 0x%x\n", | 87 | dev = sysbus_create_varargs(TYPE_STELLARIS_ADC, 0x40038000, |
79 | + offset); | 88 | qdev_get_gpio_in(nvic, 14), |
80 | } | ||
81 | } | ||
82 | |||
83 | -- | 89 | -- |
84 | 2.19.1 | 90 | 2.20.1 |
85 | 91 | ||
86 | 92 | diff view generated by jsdifflib |
1 | From: Richard Henderson <richard.henderson@linaro.org> | 1 | From: Richard Henderson <richard.henderson@linaro.org> |
---|---|---|---|
2 | 2 | ||
3 | The definition of top_bit used in this function is one higher | ||
4 | than that used in the Arm ARM psuedo-code, which put the error | ||
5 | indication at top_bit - 1 at the wrong place, which meant that | ||
6 | it wasn't visible to Auth. | ||
7 | |||
8 | Fixing the definition of top_bit requires more changes, because | ||
9 | its most common use is for the count of bits in top_bit:bot_bit, | ||
10 | which would then need to be computed as top_bit - bot_bit + 1. | ||
11 | |||
12 | For now, prefer the minimal fix to the error indication alone. | ||
13 | |||
14 | Fixes: 63ff0ca94cb | ||
15 | Reported-by: Derrick McKee <derrick.mckee@gmail.com> | ||
3 | Signed-off-by: Richard Henderson <richard.henderson@linaro.org> | 16 | Signed-off-by: Richard Henderson <richard.henderson@linaro.org> |
4 | Message-id: 20181113180154.17903-3-richard.henderson@linaro.org | 17 | Message-id: 20200728195706.11087-1-richard.henderson@linaro.org |
5 | Reviewed-by: Peter Maydell <peter.maydell@linaro.org> | 18 | Reviewed-by: Peter Maydell <peter.maydell@linaro.org> |
19 | [PMM: added comment about the divergence from the pseudocode] | ||
6 | Signed-off-by: Peter Maydell <peter.maydell@linaro.org> | 20 | Signed-off-by: Peter Maydell <peter.maydell@linaro.org> |
7 | --- | 21 | --- |
8 | target/arm/kvm64.c | 90 ++++++++++++++++++++++++++++++++++++++++++++-- | 22 | target/arm/pauth_helper.c | 6 +++++- |
9 | 1 file changed, 88 insertions(+), 2 deletions(-) | 23 | tests/tcg/aarch64/pauth-5.c | 33 +++++++++++++++++++++++++++++++ |
24 | tests/tcg/aarch64/Makefile.target | 2 +- | ||
25 | 3 files changed, 39 insertions(+), 2 deletions(-) | ||
26 | create mode 100644 tests/tcg/aarch64/pauth-5.c | ||
10 | 27 | ||
11 | diff --git a/target/arm/kvm64.c b/target/arm/kvm64.c | 28 | diff --git a/target/arm/pauth_helper.c b/target/arm/pauth_helper.c |
12 | index XXXXXXX..XXXXXXX 100644 | 29 | index XXXXXXX..XXXXXXX 100644 |
13 | --- a/target/arm/kvm64.c | 30 | --- a/target/arm/pauth_helper.c |
14 | +++ b/target/arm/kvm64.c | 31 | +++ b/target/arm/pauth_helper.c |
15 | @@ -XXX,XX +XXX,XX @@ static inline void unset_feature(uint64_t *features, int feature) | 32 | @@ -XXX,XX +XXX,XX @@ static uint64_t pauth_addpac(CPUARMState *env, uint64_t ptr, uint64_t modifier, |
16 | *features &= ~(1ULL << feature); | 33 | */ |
17 | } | 34 | test = sextract64(ptr, bot_bit, top_bit - bot_bit); |
18 | 35 | if (test != 0 && test != -1) { | |
19 | +static int read_sys_reg32(int fd, uint32_t *pret, uint64_t id) | 36 | - pac ^= MAKE_64BIT_MASK(top_bit - 1, 1); |
37 | + /* | ||
38 | + * Note that our top_bit is one greater than the pseudocode's | ||
39 | + * version, hence "- 2" here. | ||
40 | + */ | ||
41 | + pac ^= MAKE_64BIT_MASK(top_bit - 2, 1); | ||
42 | } | ||
43 | |||
44 | /* | ||
45 | diff --git a/tests/tcg/aarch64/pauth-5.c b/tests/tcg/aarch64/pauth-5.c | ||
46 | new file mode 100644 | ||
47 | index XXXXXXX..XXXXXXX | ||
48 | --- /dev/null | ||
49 | +++ b/tests/tcg/aarch64/pauth-5.c | ||
50 | @@ -XXX,XX +XXX,XX @@ | ||
51 | +#include <assert.h> | ||
52 | + | ||
53 | +static int x; | ||
54 | + | ||
55 | +int main() | ||
20 | +{ | 56 | +{ |
21 | + uint64_t ret; | 57 | + int *p0 = &x, *p1, *p2, *p3; |
22 | + struct kvm_one_reg idreg = { .id = id, .addr = (uintptr_t)&ret }; | 58 | + unsigned long salt = 0; |
23 | + int err; | ||
24 | + | 59 | + |
25 | + assert((id & KVM_REG_SIZE_MASK) == KVM_REG_SIZE_U64); | 60 | + /* |
26 | + err = ioctl(fd, KVM_GET_ONE_REG, &idreg); | 61 | + * With TBI enabled and a 48-bit VA, there are 7 bits of auth, and so |
27 | + if (err < 0) { | 62 | + * a 1/128 chance of auth = pac(ptr,key,salt) producing zero. |
28 | + return -1; | 63 | + * Find a salt that creates auth != 0. |
29 | + } | 64 | + */ |
30 | + *pret = ret; | 65 | + do { |
66 | + salt++; | ||
67 | + asm("pacda %0, %1" : "=r"(p1) : "r"(salt), "0"(p0)); | ||
68 | + } while (p0 == p1); | ||
69 | + | ||
70 | + /* | ||
71 | + * This pac must fail, because the input pointer bears an encryption, | ||
72 | + * and so is not properly extended within bits [55:47]. This will | ||
73 | + * toggle bit 54 in the output... | ||
74 | + */ | ||
75 | + asm("pacda %0, %1" : "=r"(p2) : "r"(salt), "0"(p1)); | ||
76 | + | ||
77 | + /* ... so that the aut must fail, setting bit 53 in the output ... */ | ||
78 | + asm("autda %0, %1" : "=r"(p3) : "r"(salt), "0"(p2)); | ||
79 | + | ||
80 | + /* ... which means this equality must not hold. */ | ||
81 | + assert(p3 != p0); | ||
31 | + return 0; | 82 | + return 0; |
32 | +} | 83 | +} |
33 | + | 84 | diff --git a/tests/tcg/aarch64/Makefile.target b/tests/tcg/aarch64/Makefile.target |
34 | +static int read_sys_reg64(int fd, uint64_t *pret, uint64_t id) | 85 | index XXXXXXX..XXXXXXX 100644 |
35 | +{ | 86 | --- a/tests/tcg/aarch64/Makefile.target |
36 | + struct kvm_one_reg idreg = { .id = id, .addr = (uintptr_t)pret }; | 87 | +++ b/tests/tcg/aarch64/Makefile.target |
37 | + | 88 | @@ -XXX,XX +XXX,XX @@ run-fcvt: fcvt |
38 | + assert((id & KVM_REG_SIZE_MASK) == KVM_REG_SIZE_U64); | 89 | |
39 | + return ioctl(fd, KVM_GET_ONE_REG, &idreg); | 90 | # Pauth Tests |
40 | +} | 91 | ifneq ($(DOCKER_IMAGE)$(CROSS_CC_HAS_ARMV8_3),) |
41 | + | 92 | -AARCH64_TESTS += pauth-1 pauth-2 pauth-4 |
42 | bool kvm_arm_get_host_cpu_features(ARMHostCPUFeatures *ahcf) | 93 | +AARCH64_TESTS += pauth-1 pauth-2 pauth-4 pauth-5 |
43 | { | 94 | pauth-%: CFLAGS += -march=armv8.3-a |
44 | /* Identify the feature bits corresponding to the host CPU, and | 95 | run-pauth-%: QEMU_OPTS += -cpu max |
45 | * fill out the ARMHostCPUClass fields accordingly. To do this | 96 | run-plugin-pauth-%: QEMU_OPTS += -cpu max |
46 | * we have to create a scratch VM, create a single CPU inside it, | ||
47 | * and then query that CPU for the relevant ID registers. | ||
48 | - * For AArch64 we currently don't care about ID registers at | ||
49 | - * all; we just want to know the CPU type. | ||
50 | */ | ||
51 | int fdarray[3]; | ||
52 | uint64_t features = 0; | ||
53 | + int err; | ||
54 | + | ||
55 | /* Old kernels may not know about the PREFERRED_TARGET ioctl: however | ||
56 | * we know these will only support creating one kind of guest CPU, | ||
57 | * which is its preferred CPU type. Fortunately these old kernels | ||
58 | @@ -XXX,XX +XXX,XX @@ bool kvm_arm_get_host_cpu_features(ARMHostCPUFeatures *ahcf) | ||
59 | ahcf->target = init.target; | ||
60 | ahcf->dtb_compatible = "arm,arm-v8"; | ||
61 | |||
62 | + err = read_sys_reg64(fdarray[2], &ahcf->isar.id_aa64pfr0, | ||
63 | + ARM64_SYS_REG(3, 0, 0, 4, 0)); | ||
64 | + if (unlikely(err < 0)) { | ||
65 | + /* | ||
66 | + * Before v4.15, the kernel only exposed a limited number of system | ||
67 | + * registers, not including any of the interesting AArch64 ID regs. | ||
68 | + * For the most part we could leave these fields as zero with minimal | ||
69 | + * effect, since this does not affect the values seen by the guest. | ||
70 | + * | ||
71 | + * However, it could cause problems down the line for QEMU, | ||
72 | + * so provide a minimal v8.0 default. | ||
73 | + * | ||
74 | + * ??? Could read MIDR and use knowledge from cpu64.c. | ||
75 | + * ??? Could map a page of memory into our temp guest and | ||
76 | + * run the tiniest of hand-crafted kernels to extract | ||
77 | + * the values seen by the guest. | ||
78 | + * ??? Either of these sounds like too much effort just | ||
79 | + * to work around running a modern host kernel. | ||
80 | + */ | ||
81 | + ahcf->isar.id_aa64pfr0 = 0x00000011; /* EL1&0, AArch64 only */ | ||
82 | + err = 0; | ||
83 | + } else { | ||
84 | + err |= read_sys_reg64(fdarray[2], &ahcf->isar.id_aa64pfr1, | ||
85 | + ARM64_SYS_REG(3, 0, 0, 4, 1)); | ||
86 | + err |= read_sys_reg64(fdarray[2], &ahcf->isar.id_aa64isar0, | ||
87 | + ARM64_SYS_REG(3, 0, 0, 6, 0)); | ||
88 | + err |= read_sys_reg64(fdarray[2], &ahcf->isar.id_aa64isar1, | ||
89 | + ARM64_SYS_REG(3, 0, 0, 6, 1)); | ||
90 | + | ||
91 | + /* | ||
92 | + * Note that if AArch32 support is not present in the host, | ||
93 | + * the AArch32 sysregs are present to be read, but will | ||
94 | + * return UNKNOWN values. This is neither better nor worse | ||
95 | + * than skipping the reads and leaving 0, as we must avoid | ||
96 | + * considering the values in every case. | ||
97 | + */ | ||
98 | + err |= read_sys_reg32(fdarray[2], &ahcf->isar.id_isar0, | ||
99 | + ARM64_SYS_REG(3, 0, 0, 2, 0)); | ||
100 | + err |= read_sys_reg32(fdarray[2], &ahcf->isar.id_isar1, | ||
101 | + ARM64_SYS_REG(3, 0, 0, 2, 1)); | ||
102 | + err |= read_sys_reg32(fdarray[2], &ahcf->isar.id_isar2, | ||
103 | + ARM64_SYS_REG(3, 0, 0, 2, 2)); | ||
104 | + err |= read_sys_reg32(fdarray[2], &ahcf->isar.id_isar3, | ||
105 | + ARM64_SYS_REG(3, 0, 0, 2, 3)); | ||
106 | + err |= read_sys_reg32(fdarray[2], &ahcf->isar.id_isar4, | ||
107 | + ARM64_SYS_REG(3, 0, 0, 2, 4)); | ||
108 | + err |= read_sys_reg32(fdarray[2], &ahcf->isar.id_isar5, | ||
109 | + ARM64_SYS_REG(3, 0, 0, 2, 5)); | ||
110 | + err |= read_sys_reg32(fdarray[2], &ahcf->isar.id_isar6, | ||
111 | + ARM64_SYS_REG(3, 0, 0, 2, 7)); | ||
112 | + | ||
113 | + err |= read_sys_reg32(fdarray[2], &ahcf->isar.mvfr0, | ||
114 | + ARM64_SYS_REG(3, 0, 0, 3, 0)); | ||
115 | + err |= read_sys_reg32(fdarray[2], &ahcf->isar.mvfr1, | ||
116 | + ARM64_SYS_REG(3, 0, 0, 3, 1)); | ||
117 | + err |= read_sys_reg32(fdarray[2], &ahcf->isar.mvfr2, | ||
118 | + ARM64_SYS_REG(3, 0, 0, 3, 2)); | ||
119 | + } | ||
120 | + | ||
121 | kvm_arm_destroy_scratch_host_vcpu(fdarray); | ||
122 | |||
123 | + if (err < 0) { | ||
124 | + return false; | ||
125 | + } | ||
126 | + | ||
127 | /* We can assume any KVM supporting CPU is at least a v8 | ||
128 | * with VFPv4+Neon; this in turn implies most of the other | ||
129 | * feature bits. | ||
130 | -- | 97 | -- |
131 | 2.19.1 | 98 | 2.20.1 |
132 | 99 | ||
133 | 100 | diff view generated by jsdifflib |
1 | An off-by-one error in a switch case in onenand_read() allowed | 1 | From: Kaige Li <likaige@loongson.cn> |
---|---|---|---|
2 | a misbehaving guest to read off the end of a block of memory. | ||
3 | 2 | ||
4 | NB: the onenand device is used only by the "n800" and "n810" | 3 | GCC version 4.9.4 isn't clever enough to figure out that all |
5 | machines, which are usable only with TCG, not KVM, so this is | 4 | execution paths in disas_ldst() that use 'fn' will have initialized |
6 | not a security issue. | 5 | it first, and so it warns: |
7 | 6 | ||
8 | Reported-by: Thomas Huth <thuth@redhat.com> | 7 | /home/LiKaige/qemu/target/arm/translate-a64.c: In function ‘disas_ldst’: |
9 | Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> | 8 | /home/LiKaige/qemu/target/arm/translate-a64.c:3392:5: error: ‘fn’ may be used uninitialized in this function [-Werror=maybe-uninitialized] |
10 | Reviewed-by: Richard Henderson <richard.henderson@linaro.org> | 9 | fn(cpu_reg(s, rt), clean_addr, tcg_rs, get_mem_index(s), |
11 | Signed-off-by: Peter Maydell <peter.maydell@linaro.org> | 10 | ^ |
12 | Message-id: 20181115143535.5885-2-peter.maydell@linaro.org | 11 | /home/LiKaige/qemu/target/arm/translate-a64.c:3318:22: note: ‘fn’ was declared here |
13 | Suggested-by: Richard Henderson <richard.henderson@linaro.org> | 12 | AtomicThreeOpFn *fn; |
13 | ^ | ||
14 | |||
15 | Make it happy by initializing the variable to NULL. | ||
16 | |||
17 | Signed-off-by: Kaige Li <likaige@loongson.cn> | ||
18 | Message-id: 1596110248-7366-2-git-send-email-likaige@loongson.cn | ||
19 | Reviewed-by: Peter Maydell <peter.maydell@linaro.org> | ||
20 | [PMM: Clean up commit message and note which gcc version this was] | ||
14 | Signed-off-by: Peter Maydell <peter.maydell@linaro.org> | 21 | Signed-off-by: Peter Maydell <peter.maydell@linaro.org> |
15 | --- | 22 | --- |
16 | hw/block/onenand.c | 2 +- | 23 | target/arm/translate-a64.c | 2 +- |
17 | 1 file changed, 1 insertion(+), 1 deletion(-) | 24 | 1 file changed, 1 insertion(+), 1 deletion(-) |
18 | 25 | ||
19 | diff --git a/hw/block/onenand.c b/hw/block/onenand.c | 26 | diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c |
20 | index XXXXXXX..XXXXXXX 100644 | 27 | index XXXXXXX..XXXXXXX 100644 |
21 | --- a/hw/block/onenand.c | 28 | --- a/target/arm/translate-a64.c |
22 | +++ b/hw/block/onenand.c | 29 | +++ b/target/arm/translate-a64.c |
23 | @@ -XXX,XX +XXX,XX @@ static uint64_t onenand_read(void *opaque, hwaddr addr, | 30 | @@ -XXX,XX +XXX,XX @@ static void disas_ldst_atomic(DisasContext *s, uint32_t insn, |
24 | int offset = addr >> s->shift; | 31 | bool r = extract32(insn, 22, 1); |
25 | 32 | bool a = extract32(insn, 23, 1); | |
26 | switch (offset) { | 33 | TCGv_i64 tcg_rs, clean_addr; |
27 | - case 0x0000 ... 0xc000: | 34 | - AtomicThreeOpFn *fn; |
28 | + case 0x0000 ... 0xbffe: | 35 | + AtomicThreeOpFn *fn = NULL; |
29 | return lduw_le_p(s->boot[0] + addr); | 36 | |
30 | 37 | if (is_vector || !dc_isar_feature(aa64_atomics, s)) { | |
31 | case 0xf000: /* Manufacturer ID */ | 38 | unallocated_encoding(s); |
32 | -- | 39 | -- |
33 | 2.19.1 | 40 | 2.20.1 |
34 | 41 | ||
35 | 42 | diff view generated by jsdifflib |
1 | In practice for most of the more-or-less orphan Arm board models, | 1 | The nrf51 SoC model wasn't setting the system_clock_scale |
---|---|---|---|
2 | I will review patches and put them in via the target-arm tree. | 2 | global.which meant that if guest code used the systick timer in "use |
3 | So list myself as an "Odd Fixes" status maintainer for them. | 3 | the processor clock" mode it would hang because time never advances. |
4 | 4 | ||
5 | This commit downgrades these boards to "Odd Fixes": | 5 | Set the global to match the documented CPU clock speed for this SoC. |
6 | * Allwinner-A10 | ||
7 | * Exynos | ||
8 | * Calxeda Highbank | ||
9 | * Canon DIGIC | ||
10 | * Musicpal | ||
11 | * nSeries | ||
12 | * Palm | ||
13 | * PXA2xx | ||
14 | 6 | ||
15 | These boards were already "Odd Fixes": | 7 | This SoC in fact doesn't have a SysTick timer (which is the only thing |
16 | * Gumstix | 8 | currently that cares about the system_clock_scale), because it's |
17 | * i.MX31 (kzm) | 9 | a configurable option in the Cortex-M0. However our Cortex-M0 and |
18 | 10 | thus our nrf51 and our micro:bit board do provide a SysTick, so | |
19 | Philippe Mathieu-Daudé has requested to be moved to R: | 11 | we ought to provide a functional one rather than a broken one. |
20 | status for Gumstix now that I am listed as the M: contact. | ||
21 | |||
22 | Some boards are maintained, but their patches still go | ||
23 | via the target-arm tree, so add myself as a secondary | ||
24 | maintainer contact for those: | ||
25 | * Xilinx Zynq | ||
26 | * Xilinx ZynqMP | ||
27 | * STM32F205 | ||
28 | * Netduino 2 | ||
29 | * SmartFusion2 | ||
30 | * Mecraft M2S-FG484 | ||
31 | * ASPEED BMCs | ||
32 | * NRF51 | ||
33 | 12 | ||
34 | Signed-off-by: Peter Maydell <peter.maydell@linaro.org> | 13 | Signed-off-by: Peter Maydell <peter.maydell@linaro.org> |
35 | Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> | 14 | Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> |
36 | Reviewed-by: Thomas Huth <thuth@redhat.com> | 15 | Message-id: 20200727193458.31250-1-peter.maydell@linaro.org |
37 | Message-id: 20181108134139.31666-1-peter.maydell@linaro.org | ||
38 | --- | 16 | --- |
39 | MAINTAINERS | 36 +++++++++++++++++++++++++++--------- | 17 | hw/arm/nrf51_soc.c | 5 +++++ |
40 | 1 file changed, 27 insertions(+), 9 deletions(-) | 18 | 1 file changed, 5 insertions(+) |
41 | 19 | ||
42 | diff --git a/MAINTAINERS b/MAINTAINERS | 20 | diff --git a/hw/arm/nrf51_soc.c b/hw/arm/nrf51_soc.c |
43 | index XXXXXXX..XXXXXXX 100644 | 21 | index XXXXXXX..XXXXXXX 100644 |
44 | --- a/MAINTAINERS | 22 | --- a/hw/arm/nrf51_soc.c |
45 | +++ b/MAINTAINERS | 23 | +++ b/hw/arm/nrf51_soc.c |
46 | @@ -XXX,XX +XXX,XX @@ ARM Machines | 24 | @@ -XXX,XX +XXX,XX @@ |
47 | ------------ | 25 | |
48 | Allwinner-a10 | 26 | #define BASE_TO_IRQ(base) ((base >> 12) & 0x1F) |
49 | M: Beniamino Galvani <b.galvani@gmail.com> | 27 | |
50 | +M: Peter Maydell <peter.maydell@linaro.org> | 28 | +/* HCLK (the main CPU clock) on this SoC is always 16MHz */ |
51 | L: qemu-arm@nongnu.org | 29 | +#define HCLK_FRQ 16000000 |
52 | -S: Maintained | 30 | + |
53 | +S: Odd Fixes | 31 | static uint64_t clock_read(void *opaque, hwaddr addr, unsigned int size) |
54 | F: hw/*/allwinner* | 32 | { |
55 | F: include/hw/*/allwinner* | 33 | qemu_log_mask(LOG_UNIMP, "%s: 0x%" HWADDR_PRIx " [%u]\n", |
56 | F: hw/arm/cubieboard.c | 34 | @@ -XXX,XX +XXX,XX @@ static void nrf51_soc_realize(DeviceState *dev_soc, Error **errp) |
57 | @@ -XXX,XX +XXX,XX @@ F: tests/test-arm-mptimer.c | 35 | return; |
58 | 36 | } | |
59 | Exynos | 37 | |
60 | M: Igor Mitsyanko <i.mitsyanko@gmail.com> | 38 | + system_clock_scale = NANOSECONDS_PER_SECOND / HCLK_FRQ; |
61 | +M: Peter Maydell <peter.maydell@linaro.org> | 39 | + |
62 | L: qemu-arm@nongnu.org | 40 | object_property_set_link(OBJECT(&s->cpu), "memory", OBJECT(&s->container), |
63 | -S: Maintained | 41 | &error_abort); |
64 | +S: Odd Fixes | 42 | if (!sysbus_realize(SYS_BUS_DEVICE(&s->cpu), errp)) { |
65 | F: hw/*/exynos* | ||
66 | F: include/hw/arm/exynos4210.h | ||
67 | |||
68 | Calxeda Highbank | ||
69 | M: Rob Herring <robh@kernel.org> | ||
70 | +M: Peter Maydell <peter.maydell@linaro.org> | ||
71 | L: qemu-arm@nongnu.org | ||
72 | -S: Maintained | ||
73 | +S: Odd Fixes | ||
74 | F: hw/arm/highbank.c | ||
75 | F: hw/net/xgmac.c | ||
76 | |||
77 | Canon DIGIC | ||
78 | M: Antony Pavlov <antonynpavlov@gmail.com> | ||
79 | +M: Peter Maydell <peter.maydell@linaro.org> | ||
80 | L: qemu-arm@nongnu.org | ||
81 | -S: Maintained | ||
82 | +S: Odd Fixes | ||
83 | F: include/hw/arm/digic.h | ||
84 | F: hw/*/digic* | ||
85 | |||
86 | Gumstix | ||
87 | -M: Philippe Mathieu-Daudé <f4bug@amsat.org> | ||
88 | +M: Peter Maydell <peter.maydell@linaro.org> | ||
89 | +R: Philippe Mathieu-Daudé <f4bug@amsat.org> | ||
90 | L: qemu-devel@nongnu.org | ||
91 | L: qemu-arm@nongnu.org | ||
92 | S: Odd Fixes | ||
93 | @@ -XXX,XX +XXX,XX @@ F: hw/arm/gumstix.c | ||
94 | |||
95 | i.MX31 (kzm) | ||
96 | M: Peter Chubb <peter.chubb@nicta.com.au> | ||
97 | +M: Peter Maydell <peter.maydell@linaro.org> | ||
98 | L: qemu-arm@nongnu.org | ||
99 | S: Odd Fixes | ||
100 | F: hw/arm/kzm.c | ||
101 | @@ -XXX,XX +XXX,XX @@ F: include/hw/misc/iotkit-sysinfo.h | ||
102 | |||
103 | Musicpal | ||
104 | M: Jan Kiszka <jan.kiszka@web.de> | ||
105 | +M: Peter Maydell <peter.maydell@linaro.org> | ||
106 | L: qemu-arm@nongnu.org | ||
107 | -S: Maintained | ||
108 | +S: Odd Fixes | ||
109 | F: hw/arm/musicpal.c | ||
110 | |||
111 | nSeries | ||
112 | M: Andrzej Zaborowski <balrogg@gmail.com> | ||
113 | +M: Peter Maydell <peter.maydell@linaro.org> | ||
114 | L: qemu-arm@nongnu.org | ||
115 | -S: Maintained | ||
116 | +S: Odd Fixes | ||
117 | F: hw/arm/nseries.c | ||
118 | |||
119 | Palm | ||
120 | M: Andrzej Zaborowski <balrogg@gmail.com> | ||
121 | +M: Peter Maydell <peter.maydell@linaro.org> | ||
122 | L: qemu-arm@nongnu.org | ||
123 | -S: Maintained | ||
124 | +S: Odd Fixes | ||
125 | F: hw/arm/palm.c | ||
126 | |||
127 | Raspberry Pi | ||
128 | @@ -XXX,XX +XXX,XX @@ F: include/hw/intc/realview_gic.h | ||
129 | |||
130 | PXA2XX | ||
131 | M: Andrzej Zaborowski <balrogg@gmail.com> | ||
132 | +M: Peter Maydell <peter.maydell@linaro.org> | ||
133 | L: qemu-arm@nongnu.org | ||
134 | -S: Maintained | ||
135 | +S: Odd Fixes | ||
136 | F: hw/arm/mainstone.c | ||
137 | F: hw/arm/spitz.c | ||
138 | F: hw/arm/tosa.c | ||
139 | @@ -XXX,XX +XXX,XX @@ F: include/hw/arm/virt.h | ||
140 | Xilinx Zynq | ||
141 | M: Edgar E. Iglesias <edgar.iglesias@gmail.com> | ||
142 | M: Alistair Francis <alistair@alistair23.me> | ||
143 | +M: Peter Maydell <peter.maydell@linaro.org> | ||
144 | L: qemu-arm@nongnu.org | ||
145 | S: Maintained | ||
146 | F: hw/*/xilinx_* | ||
147 | @@ -XXX,XX +XXX,XX @@ X: hw/ssi/xilinx_* | ||
148 | Xilinx ZynqMP | ||
149 | M: Alistair Francis <alistair@alistair23.me> | ||
150 | M: Edgar E. Iglesias <edgar.iglesias@gmail.com> | ||
151 | +M: Peter Maydell <peter.maydell@linaro.org> | ||
152 | L: qemu-arm@nongnu.org | ||
153 | S: Maintained | ||
154 | F: hw/*/xlnx*.c | ||
155 | @@ -XXX,XX +XXX,XX @@ F: hw/arm/virt-acpi-build.c | ||
156 | |||
157 | STM32F205 | ||
158 | M: Alistair Francis <alistair@alistair23.me> | ||
159 | +M: Peter Maydell <peter.maydell@linaro.org> | ||
160 | S: Maintained | ||
161 | F: hw/arm/stm32f205_soc.c | ||
162 | F: hw/misc/stm32f2xx_syscfg.c | ||
163 | @@ -XXX,XX +XXX,XX @@ F: include/hw/*/stm32*.h | ||
164 | |||
165 | Netduino 2 | ||
166 | M: Alistair Francis <alistair@alistair23.me> | ||
167 | +M: Peter Maydell <peter.maydell@linaro.org> | ||
168 | S: Maintained | ||
169 | F: hw/arm/netduino2.c | ||
170 | |||
171 | SmartFusion2 | ||
172 | M: Subbaraya Sundeep <sundeep.lkml@gmail.com> | ||
173 | +M: Peter Maydell <peter.maydell@linaro.org> | ||
174 | S: Maintained | ||
175 | F: hw/arm/msf2-soc.c | ||
176 | F: hw/misc/msf2-sysreg.c | ||
177 | @@ -XXX,XX +XXX,XX @@ F: include/hw/ssi/mss-spi.h | ||
178 | |||
179 | Emcraft M2S-FG484 | ||
180 | M: Subbaraya Sundeep <sundeep.lkml@gmail.com> | ||
181 | +M: Peter Maydell <peter.maydell@linaro.org> | ||
182 | S: Maintained | ||
183 | F: hw/arm/msf2-som.c | ||
184 | |||
185 | ASPEED BMCs | ||
186 | M: Cédric Le Goater <clg@kaod.org> | ||
187 | +M: Peter Maydell <peter.maydell@linaro.org> | ||
188 | R: Andrew Jeffery <andrew@aj.id.au> | ||
189 | R: Joel Stanley <joel@jms.id.au> | ||
190 | L: qemu-arm@nongnu.org | ||
191 | @@ -XXX,XX +XXX,XX @@ F: include/hw/net/ftgmac100.h | ||
192 | |||
193 | NRF51 | ||
194 | M: Joel Stanley <joel@jms.id.au> | ||
195 | +M: Peter Maydell <peter.maydell@linaro.org> | ||
196 | L: qemu-arm@nongnu.org | ||
197 | S: Maintained | ||
198 | F: hw/arm/nrf51_soc.c | ||
199 | -- | 43 | -- |
200 | 2.19.1 | 44 | 2.20.1 |
201 | 45 | ||
202 | 46 | diff view generated by jsdifflib |
1 | From: Richard Henderson <richard.henderson@linaro.org> | 1 | The imx_epit device has a software-controllable reset triggered by |
---|---|---|---|
2 | setting the SWR bit in the CR register. An error in commit cc2722ec83ad9 | ||
3 | means that we will end up assert()ing if the guest does this, because | ||
4 | the code in imx_epit_write() starts ptimer transactions, and then | ||
5 | imx_epit_reset() also starts ptimer transactions, triggering | ||
6 | "ptimer_transaction_begin: Assertion `!s->in_transaction' failed". | ||
2 | 7 | ||
3 | The ID registers are replacing (some of) the feature bits. | 8 | The cleanest way to avoid this double-transaction is to move the |
4 | We need (some of) these values to determine the set of data | 9 | start-transaction for the CR write handling down below the check of |
5 | to be handled during migration. | 10 | the SWR bit. |
6 | 11 | ||
7 | Signed-off-by: Richard Henderson <richard.henderson@linaro.org> | 12 | Fixes: https://bugs.launchpad.net/qemu/+bug/1880424 |
8 | Message-id: 20181113180154.17903-2-richard.henderson@linaro.org | 13 | Fixes: cc2722ec83ad944505fe |
9 | Reviewed-by: Peter Maydell <peter.maydell@linaro.org> | ||
10 | Signed-off-by: Peter Maydell <peter.maydell@linaro.org> | 14 | Signed-off-by: Peter Maydell <peter.maydell@linaro.org> |
15 | Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> | ||
16 | Message-id: 20200727154550.3409-1-peter.maydell@linaro.org | ||
11 | --- | 17 | --- |
12 | target/arm/kvm_arm.h | 1 + | 18 | hw/timer/imx_epit.c | 13 ++++++++++--- |
13 | target/arm/kvm.c | 1 + | 19 | 1 file changed, 10 insertions(+), 3 deletions(-) |
14 | 2 files changed, 2 insertions(+) | ||
15 | 20 | ||
16 | diff --git a/target/arm/kvm_arm.h b/target/arm/kvm_arm.h | 21 | diff --git a/hw/timer/imx_epit.c b/hw/timer/imx_epit.c |
17 | index XXXXXXX..XXXXXXX 100644 | 22 | index XXXXXXX..XXXXXXX 100644 |
18 | --- a/target/arm/kvm_arm.h | 23 | --- a/hw/timer/imx_epit.c |
19 | +++ b/target/arm/kvm_arm.h | 24 | +++ b/hw/timer/imx_epit.c |
20 | @@ -XXX,XX +XXX,XX @@ void kvm_arm_destroy_scratch_host_vcpu(int *fdarray); | 25 | @@ -XXX,XX +XXX,XX @@ static void imx_epit_write(void *opaque, hwaddr offset, uint64_t value, |
21 | * by asking the host kernel) | 26 | |
22 | */ | 27 | switch (offset >> 2) { |
23 | typedef struct ARMHostCPUFeatures { | 28 | case 0: /* CR */ |
24 | + ARMISARegisters isar; | 29 | - ptimer_transaction_begin(s->timer_cmp); |
25 | uint64_t features; | 30 | - ptimer_transaction_begin(s->timer_reload); |
26 | uint32_t target; | 31 | |
27 | const char *dtb_compatible; | 32 | oldcr = s->cr; |
28 | diff --git a/target/arm/kvm.c b/target/arm/kvm.c | 33 | s->cr = value & 0x03ffffff; |
29 | index XXXXXXX..XXXXXXX 100644 | 34 | if (s->cr & CR_SWR) { |
30 | --- a/target/arm/kvm.c | 35 | /* handle the reset */ |
31 | +++ b/target/arm/kvm.c | 36 | imx_epit_reset(DEVICE(s)); |
32 | @@ -XXX,XX +XXX,XX @@ void kvm_arm_set_cpu_features_from_host(ARMCPU *cpu) | 37 | - } else { |
33 | 38 | + /* | |
34 | cpu->kvm_target = arm_host_cpu_features.target; | 39 | + * TODO: could we 'break' here? following operations appear |
35 | cpu->dtb_compatible = arm_host_cpu_features.dtb_compatible; | 40 | + * to duplicate the work imx_epit_reset() already did. |
36 | + cpu->isar = arm_host_cpu_features.isar; | 41 | + */ |
37 | env->features = arm_host_cpu_features.features; | 42 | + } |
38 | } | 43 | + |
44 | + ptimer_transaction_begin(s->timer_cmp); | ||
45 | + ptimer_transaction_begin(s->timer_reload); | ||
46 | + | ||
47 | + if (!(s->cr & CR_SWR)) { | ||
48 | imx_epit_set_freq(s); | ||
49 | } | ||
39 | 50 | ||
40 | -- | 51 | -- |
41 | 2.19.1 | 52 | 2.20.1 |
42 | 53 | ||
43 | 54 | diff view generated by jsdifflib |
Deleted patch | |||
---|---|---|---|
1 | From: Richard Henderson <richard.henderson@linaro.org> | ||
2 | 1 | ||
3 | Signed-off-by: Richard Henderson <richard.henderson@linaro.org> | ||
4 | Message-id: 20181113180154.17903-5-richard.henderson@linaro.org | ||
5 | Reviewed-by: Peter Maydell <peter.maydell@linaro.org> | ||
6 | Signed-off-by: Peter Maydell <peter.maydell@linaro.org> | ||
7 | --- | ||
8 | target/arm/kvm32.c | 40 +++++++++++++++++++++++++++++++++++----- | ||
9 | 1 file changed, 35 insertions(+), 5 deletions(-) | ||
10 | |||
11 | diff --git a/target/arm/kvm32.c b/target/arm/kvm32.c | ||
12 | index XXXXXXX..XXXXXXX 100644 | ||
13 | --- a/target/arm/kvm32.c | ||
14 | +++ b/target/arm/kvm32.c | ||
15 | @@ -XXX,XX +XXX,XX @@ bool kvm_arm_get_host_cpu_features(ARMHostCPUFeatures *ahcf) | ||
16 | * and then query that CPU for the relevant ID registers. | ||
17 | */ | ||
18 | int err = 0, fdarray[3]; | ||
19 | - uint32_t midr, id_pfr0, mvfr1; | ||
20 | + uint32_t midr, id_pfr0; | ||
21 | uint64_t features = 0; | ||
22 | |||
23 | /* Old kernels may not know about the PREFERRED_TARGET ioctl: however | ||
24 | @@ -XXX,XX +XXX,XX @@ bool kvm_arm_get_host_cpu_features(ARMHostCPUFeatures *ahcf) | ||
25 | |||
26 | err |= read_sys_reg32(fdarray[2], &midr, ARM_CP15_REG32(0, 0, 0, 0)); | ||
27 | err |= read_sys_reg32(fdarray[2], &id_pfr0, ARM_CP15_REG32(0, 0, 1, 0)); | ||
28 | - err |= read_sys_reg32(fdarray[2], &mvfr1, | ||
29 | + | ||
30 | + err |= read_sys_reg32(fdarray[2], &ahcf->isar.id_isar0, | ||
31 | + ARM_CP15_REG32(0, 0, 2, 0)); | ||
32 | + err |= read_sys_reg32(fdarray[2], &ahcf->isar.id_isar1, | ||
33 | + ARM_CP15_REG32(0, 0, 2, 1)); | ||
34 | + err |= read_sys_reg32(fdarray[2], &ahcf->isar.id_isar2, | ||
35 | + ARM_CP15_REG32(0, 0, 2, 2)); | ||
36 | + err |= read_sys_reg32(fdarray[2], &ahcf->isar.id_isar3, | ||
37 | + ARM_CP15_REG32(0, 0, 2, 3)); | ||
38 | + err |= read_sys_reg32(fdarray[2], &ahcf->isar.id_isar4, | ||
39 | + ARM_CP15_REG32(0, 0, 2, 4)); | ||
40 | + err |= read_sys_reg32(fdarray[2], &ahcf->isar.id_isar5, | ||
41 | + ARM_CP15_REG32(0, 0, 2, 5)); | ||
42 | + if (read_sys_reg32(fdarray[2], &ahcf->isar.id_isar6, | ||
43 | + ARM_CP15_REG32(0, 0, 2, 7))) { | ||
44 | + /* | ||
45 | + * Older kernels don't support reading ID_ISAR6. This register was | ||
46 | + * only introduced in ARMv8, so we can assume that it is zero on a | ||
47 | + * CPU that a kernel this old is running on. | ||
48 | + */ | ||
49 | + ahcf->isar.id_isar6 = 0; | ||
50 | + } | ||
51 | + | ||
52 | + err |= read_sys_reg32(fdarray[2], &ahcf->isar.mvfr0, | ||
53 | + KVM_REG_ARM | KVM_REG_SIZE_U32 | | ||
54 | + KVM_REG_ARM_VFP | KVM_REG_ARM_VFP_MVFR0); | ||
55 | + err |= read_sys_reg32(fdarray[2], &ahcf->isar.mvfr1, | ||
56 | KVM_REG_ARM | KVM_REG_SIZE_U32 | | ||
57 | KVM_REG_ARM_VFP | KVM_REG_ARM_VFP_MVFR1); | ||
58 | + /* | ||
59 | + * FIXME: There is not yet a way to read MVFR2. | ||
60 | + * Fortunately there is not yet anything in there that affects migration. | ||
61 | + */ | ||
62 | |||
63 | kvm_arm_destroy_scratch_host_vcpu(fdarray); | ||
64 | |||
65 | @@ -XXX,XX +XXX,XX @@ bool kvm_arm_get_host_cpu_features(ARMHostCPUFeatures *ahcf) | ||
66 | if (extract32(id_pfr0, 12, 4) == 1) { | ||
67 | set_feature(&features, ARM_FEATURE_THUMB2EE); | ||
68 | } | ||
69 | - if (extract32(mvfr1, 20, 4) == 1) { | ||
70 | + if (extract32(ahcf->isar.mvfr1, 20, 4) == 1) { | ||
71 | set_feature(&features, ARM_FEATURE_VFP_FP16); | ||
72 | } | ||
73 | - if (extract32(mvfr1, 12, 4) == 1) { | ||
74 | + if (extract32(ahcf->isar.mvfr1, 12, 4) == 1) { | ||
75 | set_feature(&features, ARM_FEATURE_NEON); | ||
76 | } | ||
77 | - if (extract32(mvfr1, 28, 4) == 1) { | ||
78 | + if (extract32(ahcf->isar.mvfr1, 28, 4) == 1) { | ||
79 | /* FMAC support implies VFPv4 */ | ||
80 | set_feature(&features, ARM_FEATURE_VFP4); | ||
81 | } | ||
82 | -- | ||
83 | 2.19.1 | ||
84 | |||
85 | diff view generated by jsdifflib |
Deleted patch | |||
---|---|---|---|
1 | From: Thomas Huth <thuth@redhat.com> | ||
2 | 1 | ||
3 | Add entries for the boards "mcimx6ul-evk", "mcimx7d-sabre", "raspi2", | ||
4 | "raspi3", "sabrelite", "vexpress-a15", "vexpress-a9" and "virt". | ||
5 | While we're at it, also adjust the "i.MX31" section a little bit, | ||
6 | so that the wildcards there do not match anymore for unrelated files | ||
7 | (e.g. the new hw/misc/imx6ul_ccm.c file). | ||
8 | |||
9 | Signed-off-by: Thomas Huth <thuth@redhat.com> | ||
10 | Message-id: 1542184999-11145-1-git-send-email-thuth@redhat.com | ||
11 | Signed-off-by: Peter Maydell <peter.maydell@linaro.org> | ||
12 | --- | ||
13 | MAINTAINERS | 70 +++++++++++++++++++++++++++++++++++++++++++++++++---- | ||
14 | 1 file changed, 65 insertions(+), 5 deletions(-) | ||
15 | |||
16 | diff --git a/MAINTAINERS b/MAINTAINERS | ||
17 | index XXXXXXX..XXXXXXX 100644 | ||
18 | --- a/MAINTAINERS | ||
19 | +++ b/MAINTAINERS | ||
20 | @@ -XXX,XX +XXX,XX @@ L: qemu-arm@nongnu.org | ||
21 | S: Odd Fixes | ||
22 | F: hw/arm/gumstix.c | ||
23 | |||
24 | -i.MX31 | ||
25 | +i.MX31 (kzm) | ||
26 | M: Peter Chubb <peter.chubb@nicta.com.au> | ||
27 | L: qemu-arm@nongnu.org | ||
28 | -S: Odd fixes | ||
29 | -F: hw/*/imx* | ||
30 | -F: include/hw/*/imx* | ||
31 | +S: Odd Fixes | ||
32 | F: hw/arm/kzm.c | ||
33 | -F: include/hw/arm/fsl-imx31.h | ||
34 | +F: hw/*/imx_* | ||
35 | +F: hw/*/*imx31* | ||
36 | +F: include/hw/*/imx_* | ||
37 | +F: include/hw/*/*imx31* | ||
38 | |||
39 | Integrator CP | ||
40 | M: Peter Maydell <peter.maydell@linaro.org> | ||
41 | @@ -XXX,XX +XXX,XX @@ S: Maintained | ||
42 | F: hw/arm/integratorcp.c | ||
43 | F: hw/misc/arm_integrator_debug.c | ||
44 | |||
45 | +MCIMX6UL EVK / i.MX6ul | ||
46 | +M: Peter Maydell <peter.maydell@linaro.org> | ||
47 | +R: Jean-Christophe Dubois <jcd@tribudubois.net> | ||
48 | +L: qemu-arm@nongnu.org | ||
49 | +S: Odd Fixes | ||
50 | +F: hw/arm/mcimx6ul-evk.c | ||
51 | +F: hw/arm/fsl-imx6ul.c | ||
52 | +F: hw/misc/imx6ul_ccm.c | ||
53 | +F: include/hw/arm/fsl-imx6ul.h | ||
54 | +F: include/hw/misc/imx6ul_ccm.h | ||
55 | + | ||
56 | +MCIMX7D SABRE / i.MX7 | ||
57 | +M: Peter Maydell <peter.maydell@linaro.org> | ||
58 | +R: Andrey Smirnov <andrew.smirnov@gmail.com> | ||
59 | +L: qemu-arm@nongnu.org | ||
60 | +S: Odd Fixes | ||
61 | +F: hw/arm/mcimx7d-sabre.c | ||
62 | +F: hw/arm/fsl-imx7.c | ||
63 | +F: include/hw/arm/fsl-imx7.h | ||
64 | +F: hw/pci-host/designware.c | ||
65 | +F: include/hw/pci-host/designware.h | ||
66 | + | ||
67 | MPS2 | ||
68 | M: Peter Maydell <peter.maydell@linaro.org> | ||
69 | L: qemu-arm@nongnu.org | ||
70 | @@ -XXX,XX +XXX,XX @@ L: qemu-arm@nongnu.org | ||
71 | S: Maintained | ||
72 | F: hw/arm/palm.c | ||
73 | |||
74 | +Raspberry Pi | ||
75 | +M: Peter Maydell <peter.maydell@linaro.org> | ||
76 | +R: Andrew Baumann <Andrew.Baumann@microsoft.com> | ||
77 | +R: Philippe Mathieu-Daudé <f4bug@amsat.org> | ||
78 | +L: qemu-arm@nongnu.org | ||
79 | +S: Odd Fixes | ||
80 | +F: hw/arm/raspi_platform.h | ||
81 | +F: hw/*/bcm283* | ||
82 | +F: include/hw/arm/raspi* | ||
83 | +F: include/hw/*/bcm283* | ||
84 | + | ||
85 | Real View | ||
86 | M: Peter Maydell <peter.maydell@linaro.org> | ||
87 | L: qemu-arm@nongnu.org | ||
88 | @@ -XXX,XX +XXX,XX @@ F: hw/*/pxa2xx* | ||
89 | F: hw/misc/mst_fpga.c | ||
90 | F: include/hw/arm/pxa.h | ||
91 | |||
92 | +SABRELITE / i.MX6 | ||
93 | +M: Peter Maydell <peter.maydell@linaro.org> | ||
94 | +R: Jean-Christophe Dubois <jcd@tribudubois.net> | ||
95 | +L: qemu-arm@nongnu.org | ||
96 | +S: Odd Fixes | ||
97 | +F: hw/arm/sabrelite.c | ||
98 | +F: hw/arm/fsl-imx6.c | ||
99 | +F: hw/misc/imx6_src.c | ||
100 | +F: hw/ssi/imx_spi.c | ||
101 | +F: include/hw/arm/fsl-imx6.h | ||
102 | +F: include/hw/misc/imx6_src.h | ||
103 | +F: include/hw/ssi/imx_spi.h | ||
104 | + | ||
105 | Sharp SL-5500 (Collie) PDA | ||
106 | M: Peter Maydell <peter.maydell@linaro.org> | ||
107 | L: qemu-arm@nongnu.org | ||
108 | @@ -XXX,XX +XXX,XX @@ L: qemu-arm@nongnu.org | ||
109 | S: Maintained | ||
110 | F: hw/*/stellaris* | ||
111 | |||
112 | +Versatile Express | ||
113 | +M: Peter Maydell <peter.maydell@linaro.org> | ||
114 | +L: qemu-arm@nongnu.org | ||
115 | +S: Maintained | ||
116 | +F: hw/arm/vexpress.c | ||
117 | + | ||
118 | Versatile PB | ||
119 | M: Peter Maydell <peter.maydell@linaro.org> | ||
120 | L: qemu-arm@nongnu.org | ||
121 | @@ -XXX,XX +XXX,XX @@ S: Maintained | ||
122 | F: hw/*/versatile* | ||
123 | F: hw/misc/arm_sysctl.c | ||
124 | |||
125 | +Virt | ||
126 | +M: Peter Maydell <peter.maydell@linaro.org> | ||
127 | +L: qemu-arm@nongnu.org | ||
128 | +S: Maintained | ||
129 | +F: hw/arm/virt* | ||
130 | +F: include/hw/arm/virt.h | ||
131 | + | ||
132 | Xilinx Zynq | ||
133 | M: Edgar E. Iglesias <edgar.iglesias@gmail.com> | ||
134 | M: Alistair Francis <alistair@alistair23.me> | ||
135 | -- | ||
136 | 2.19.1 | ||
137 | |||
138 | diff view generated by jsdifflib |