linux/drivers/firmware/qcom_scm.c

1570 lines
39 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-only
/* Copyright (c) 2010,2015,2019 The Linux Foundation. All rights reserved.
* Copyright (C) 2015 Linaro Ltd.
*/
#include <linux/platform_device.h>
#include <linux/init.h>
firmware: qcom: scm: Add wait-queue handling logic When the firmware (FW) supports multiple requests per VM, multiple requests from the same/different VM can reach the firmware at the same time. Since the firmware currently being used has limited resources, it guards them with a resource lock and puts requests on a wait-queue internally and signals to HLOS that it is doing so. It does this by returning a new return value in addition to success or error: SCM_WAITQ_SLEEP. A sleeping SCM call can be woken up by an interrupt that the FW raises. 1) SCM_WAITQ_SLEEP: When an SCM call receives this return value instead of success or error, FW has placed this call on a wait-queue and has signalled HLOS to put it to non-interruptible sleep. Along with this return value, FW also passes to HLOS `wq_ctx` - a unique number (UID) identifying the wait-queue that it has put the call on, internally. This is to help HLOS with its own bookkeeping to wake this sleeping call later. Additionally, FW also passes to HLOS `smc_call_ctx` - a UID identifying the SCM call thus being put to sleep. This is also for HLOS' bookkeeping to wake this call up later. These two additional values are passed via the a1 and a2 registers. N.B.: The "ctx" in the above UID names = "context". The handshake mechanism that HLOS uses to talk to FW about wait-queue operations involves two new SMC calls. 1) get_wq_ctx(): Arguments: None Returns: wq_ctx, flags, more_pending Get the wait-queue context, and wake up either one or all of the sleeping SCM calls associated with that wait-queue. Additionally, repeat this if there are more wait-queues that are ready to have their requests woken up (`more_pending`). 2) wq_resume(smc_call_ctx): Arguments: smc_call_ctx HLOS needs to issue this in response to receiving an IRQ, passing to FW the same smc_call_ctx that FW receives from HLOS via the get_wq_ctx() call. (The mechanism to wake a SMC call back up is described in detail below) VM_1 VM_2 Firmware │ │ │ │ │ │ │ │ │ │ │ │ │ REQUEST_1 │ │ ├────────────────────────┼─────────────────────────────────┤ │ │ │ │ │ ┌──┼──┐ │ │ │ │ │ │ │ REQUEST_2 │ │ │ │ ├──────────────────────────────┼──┤ │ │ │ │ │ │Resource │ │ │ │ │is busy │ │ {WQ_SLEEP} │ │ │ │ │◄─────────────────────────────┼──┤ │ │ │ wq_ctx, smc_call_ctx │ │ │ │ │ └──┼──┘ │ REQUEST_1 COMPLETE │ │ │◄───────────────────────┼─────────────────────────────────┤ │ │ │ │ │ IRQ │ │ │◄─-------------------------------│ │ │ │ │ │ get_wq_ctx() │ │ ├────────────────────────────────►│ │ │ │ │ │ │ │ │◄────────────────────────────────┤ │ │ wq_ctx, flags, and │ │ │ more_pending │ │ │ │ │ │ │ │ │ wq_resume(smc_call_ctx) │ │ ├────────────────────────────────►│ │ │ │ │ │ │ │ │ REQUEST_2 COMPLETE │ │ │◄────────────────────────────────┤ │ │ │ │ │ │ With the exception of get_wq_ctx(), the other SMC call wq_resume() can return WQ_SLEEP (these nested rounds of WQ_SLEEP are not shown in the above diagram for the sake of simplicity). Therefore, introduce a new do-while loop to handle multiple WQ_SLEEP return values for the same parent SCM call. Request Completion in the above diagram refers to either a success return value (zero) or error (and not SMC_WAITQ_SLEEP) Also add the interrupt handler that wakes up a sleeping SCM call. Signed-off-by: Guru Das Srinagesh <quic_gurus@quicinc.com> Co-developed-by: Sibi Sankar <quic_sibis@quicinc.com> Signed-off-by: Sibi Sankar <quic_sibis@quicinc.com> Reviewed-by: Guru Das Srinagesh <quic_gurus@quicinc.com> Signed-off-by: Bjorn Andersson <andersson@kernel.org> Link: https://lore.kernel.org/r/20230113161114.22607-3-quic_sibis@quicinc.com
2023-01-14 00:11:14 +08:00
#include <linux/interrupt.h>
#include <linux/completion.h>
#include <linux/cpumask.h>
#include <linux/export.h>
#include <linux/dma-mapping.h>
#include <linux/interconnect.h>
#include <linux/module.h>
#include <linux/types.h>
#include <linux/firmware/qcom/qcom_scm.h>
#include <linux/of.h>
#include <linux/of_address.h>
firmware: qcom: scm: Add wait-queue handling logic When the firmware (FW) supports multiple requests per VM, multiple requests from the same/different VM can reach the firmware at the same time. Since the firmware currently being used has limited resources, it guards them with a resource lock and puts requests on a wait-queue internally and signals to HLOS that it is doing so. It does this by returning a new return value in addition to success or error: SCM_WAITQ_SLEEP. A sleeping SCM call can be woken up by an interrupt that the FW raises. 1) SCM_WAITQ_SLEEP: When an SCM call receives this return value instead of success or error, FW has placed this call on a wait-queue and has signalled HLOS to put it to non-interruptible sleep. Along with this return value, FW also passes to HLOS `wq_ctx` - a unique number (UID) identifying the wait-queue that it has put the call on, internally. This is to help HLOS with its own bookkeeping to wake this sleeping call later. Additionally, FW also passes to HLOS `smc_call_ctx` - a UID identifying the SCM call thus being put to sleep. This is also for HLOS' bookkeeping to wake this call up later. These two additional values are passed via the a1 and a2 registers. N.B.: The "ctx" in the above UID names = "context". The handshake mechanism that HLOS uses to talk to FW about wait-queue operations involves two new SMC calls. 1) get_wq_ctx(): Arguments: None Returns: wq_ctx, flags, more_pending Get the wait-queue context, and wake up either one or all of the sleeping SCM calls associated with that wait-queue. Additionally, repeat this if there are more wait-queues that are ready to have their requests woken up (`more_pending`). 2) wq_resume(smc_call_ctx): Arguments: smc_call_ctx HLOS needs to issue this in response to receiving an IRQ, passing to FW the same smc_call_ctx that FW receives from HLOS via the get_wq_ctx() call. (The mechanism to wake a SMC call back up is described in detail below) VM_1 VM_2 Firmware │ │ │ │ │ │ │ │ │ │ │ │ │ REQUEST_1 │ │ ├────────────────────────┼─────────────────────────────────┤ │ │ │ │ │ ┌──┼──┐ │ │ │ │ │ │ │ REQUEST_2 │ │ │ │ ├──────────────────────────────┼──┤ │ │ │ │ │ │Resource │ │ │ │ │is busy │ │ {WQ_SLEEP} │ │ │ │ │◄─────────────────────────────┼──┤ │ │ │ wq_ctx, smc_call_ctx │ │ │ │ │ └──┼──┘ │ REQUEST_1 COMPLETE │ │ │◄───────────────────────┼─────────────────────────────────┤ │ │ │ │ │ IRQ │ │ │◄─-------------------------------│ │ │ │ │ │ get_wq_ctx() │ │ ├────────────────────────────────►│ │ │ │ │ │ │ │ │◄────────────────────────────────┤ │ │ wq_ctx, flags, and │ │ │ more_pending │ │ │ │ │ │ │ │ │ wq_resume(smc_call_ctx) │ │ ├────────────────────────────────►│ │ │ │ │ │ │ │ │ REQUEST_2 COMPLETE │ │ │◄────────────────────────────────┤ │ │ │ │ │ │ With the exception of get_wq_ctx(), the other SMC call wq_resume() can return WQ_SLEEP (these nested rounds of WQ_SLEEP are not shown in the above diagram for the sake of simplicity). Therefore, introduce a new do-while loop to handle multiple WQ_SLEEP return values for the same parent SCM call. Request Completion in the above diagram refers to either a success return value (zero) or error (and not SMC_WAITQ_SLEEP) Also add the interrupt handler that wakes up a sleeping SCM call. Signed-off-by: Guru Das Srinagesh <quic_gurus@quicinc.com> Co-developed-by: Sibi Sankar <quic_sibis@quicinc.com> Signed-off-by: Sibi Sankar <quic_sibis@quicinc.com> Reviewed-by: Guru Das Srinagesh <quic_gurus@quicinc.com> Signed-off-by: Bjorn Andersson <andersson@kernel.org> Link: https://lore.kernel.org/r/20230113161114.22607-3-quic_sibis@quicinc.com
2023-01-14 00:11:14 +08:00
#include <linux/of_irq.h>
#include <linux/of_platform.h>
#include <linux/clk.h>
#include <linux/reset-controller.h>
#include <linux/arm-smccc.h>
#include "qcom_scm.h"
static bool download_mode = IS_ENABLED(CONFIG_QCOM_SCM_DOWNLOAD_MODE_DEFAULT);
module_param(download_mode, bool, 0);
#define SCM_HAS_CORE_CLK BIT(0)
#define SCM_HAS_IFACE_CLK BIT(1)
#define SCM_HAS_BUS_CLK BIT(2)
struct qcom_scm {
struct device *dev;
struct clk *core_clk;
struct clk *iface_clk;
struct clk *bus_clk;
struct icc_path *path;
firmware: qcom: scm: Add wait-queue handling logic When the firmware (FW) supports multiple requests per VM, multiple requests from the same/different VM can reach the firmware at the same time. Since the firmware currently being used has limited resources, it guards them with a resource lock and puts requests on a wait-queue internally and signals to HLOS that it is doing so. It does this by returning a new return value in addition to success or error: SCM_WAITQ_SLEEP. A sleeping SCM call can be woken up by an interrupt that the FW raises. 1) SCM_WAITQ_SLEEP: When an SCM call receives this return value instead of success or error, FW has placed this call on a wait-queue and has signalled HLOS to put it to non-interruptible sleep. Along with this return value, FW also passes to HLOS `wq_ctx` - a unique number (UID) identifying the wait-queue that it has put the call on, internally. This is to help HLOS with its own bookkeeping to wake this sleeping call later. Additionally, FW also passes to HLOS `smc_call_ctx` - a UID identifying the SCM call thus being put to sleep. This is also for HLOS' bookkeeping to wake this call up later. These two additional values are passed via the a1 and a2 registers. N.B.: The "ctx" in the above UID names = "context". The handshake mechanism that HLOS uses to talk to FW about wait-queue operations involves two new SMC calls. 1) get_wq_ctx(): Arguments: None Returns: wq_ctx, flags, more_pending Get the wait-queue context, and wake up either one or all of the sleeping SCM calls associated with that wait-queue. Additionally, repeat this if there are more wait-queues that are ready to have their requests woken up (`more_pending`). 2) wq_resume(smc_call_ctx): Arguments: smc_call_ctx HLOS needs to issue this in response to receiving an IRQ, passing to FW the same smc_call_ctx that FW receives from HLOS via the get_wq_ctx() call. (The mechanism to wake a SMC call back up is described in detail below) VM_1 VM_2 Firmware │ │ │ │ │ │ │ │ │ │ │ │ │ REQUEST_1 │ │ ├────────────────────────┼─────────────────────────────────┤ │ │ │ │ │ ┌──┼──┐ │ │ │ │ │ │ │ REQUEST_2 │ │ │ │ ├──────────────────────────────┼──┤ │ │ │ │ │ │Resource │ │ │ │ │is busy │ │ {WQ_SLEEP} │ │ │ │ │◄─────────────────────────────┼──┤ │ │ │ wq_ctx, smc_call_ctx │ │ │ │ │ └──┼──┘ │ REQUEST_1 COMPLETE │ │ │◄───────────────────────┼─────────────────────────────────┤ │ │ │ │ │ IRQ │ │ │◄─-------------------------------│ │ │ │ │ │ get_wq_ctx() │ │ ├────────────────────────────────►│ │ │ │ │ │ │ │ │◄────────────────────────────────┤ │ │ wq_ctx, flags, and │ │ │ more_pending │ │ │ │ │ │ │ │ │ wq_resume(smc_call_ctx) │ │ ├────────────────────────────────►│ │ │ │ │ │ │ │ │ REQUEST_2 COMPLETE │ │ │◄────────────────────────────────┤ │ │ │ │ │ │ With the exception of get_wq_ctx(), the other SMC call wq_resume() can return WQ_SLEEP (these nested rounds of WQ_SLEEP are not shown in the above diagram for the sake of simplicity). Therefore, introduce a new do-while loop to handle multiple WQ_SLEEP return values for the same parent SCM call. Request Completion in the above diagram refers to either a success return value (zero) or error (and not SMC_WAITQ_SLEEP) Also add the interrupt handler that wakes up a sleeping SCM call. Signed-off-by: Guru Das Srinagesh <quic_gurus@quicinc.com> Co-developed-by: Sibi Sankar <quic_sibis@quicinc.com> Signed-off-by: Sibi Sankar <quic_sibis@quicinc.com> Reviewed-by: Guru Das Srinagesh <quic_gurus@quicinc.com> Signed-off-by: Bjorn Andersson <andersson@kernel.org> Link: https://lore.kernel.org/r/20230113161114.22607-3-quic_sibis@quicinc.com
2023-01-14 00:11:14 +08:00
struct completion waitq_comp;
struct reset_controller_dev reset;
/* control access to the interconnect path */
struct mutex scm_bw_lock;
int scm_vote_count;
u64 dload_mode_addr;
};
struct qcom_scm_current_perm_info {
__le32 vmid;
__le32 perm;
__le64 ctx;
__le32 ctx_size;
__le32 unused;
};
struct qcom_scm_mem_map_info {
__le64 mem_addr;
__le64 mem_size;
};
/* Each bit configures cold/warm boot address for one of the 4 CPUs */
static const u8 qcom_scm_cpu_cold_bits[QCOM_SCM_BOOT_MAX_CPUS] = {
0, BIT(0), BIT(3), BIT(5)
};
static const u8 qcom_scm_cpu_warm_bits[QCOM_SCM_BOOT_MAX_CPUS] = {
BIT(2), BIT(1), BIT(4), BIT(6)
};
firmware: qcom: scm: Add wait-queue handling logic When the firmware (FW) supports multiple requests per VM, multiple requests from the same/different VM can reach the firmware at the same time. Since the firmware currently being used has limited resources, it guards them with a resource lock and puts requests on a wait-queue internally and signals to HLOS that it is doing so. It does this by returning a new return value in addition to success or error: SCM_WAITQ_SLEEP. A sleeping SCM call can be woken up by an interrupt that the FW raises. 1) SCM_WAITQ_SLEEP: When an SCM call receives this return value instead of success or error, FW has placed this call on a wait-queue and has signalled HLOS to put it to non-interruptible sleep. Along with this return value, FW also passes to HLOS `wq_ctx` - a unique number (UID) identifying the wait-queue that it has put the call on, internally. This is to help HLOS with its own bookkeeping to wake this sleeping call later. Additionally, FW also passes to HLOS `smc_call_ctx` - a UID identifying the SCM call thus being put to sleep. This is also for HLOS' bookkeeping to wake this call up later. These two additional values are passed via the a1 and a2 registers. N.B.: The "ctx" in the above UID names = "context". The handshake mechanism that HLOS uses to talk to FW about wait-queue operations involves two new SMC calls. 1) get_wq_ctx(): Arguments: None Returns: wq_ctx, flags, more_pending Get the wait-queue context, and wake up either one or all of the sleeping SCM calls associated with that wait-queue. Additionally, repeat this if there are more wait-queues that are ready to have their requests woken up (`more_pending`). 2) wq_resume(smc_call_ctx): Arguments: smc_call_ctx HLOS needs to issue this in response to receiving an IRQ, passing to FW the same smc_call_ctx that FW receives from HLOS via the get_wq_ctx() call. (The mechanism to wake a SMC call back up is described in detail below) VM_1 VM_2 Firmware │ │ │ │ │ │ │ │ │ │ │ │ │ REQUEST_1 │ │ ├────────────────────────┼─────────────────────────────────┤ │ │ │ │ │ ┌──┼──┐ │ │ │ │ │ │ │ REQUEST_2 │ │ │ │ ├──────────────────────────────┼──┤ │ │ │ │ │ │Resource │ │ │ │ │is busy │ │ {WQ_SLEEP} │ │ │ │ │◄─────────────────────────────┼──┤ │ │ │ wq_ctx, smc_call_ctx │ │ │ │ │ └──┼──┘ │ REQUEST_1 COMPLETE │ │ │◄───────────────────────┼─────────────────────────────────┤ │ │ │ │ │ IRQ │ │ │◄─-------------------------------│ │ │ │ │ │ get_wq_ctx() │ │ ├────────────────────────────────►│ │ │ │ │ │ │ │ │◄────────────────────────────────┤ │ │ wq_ctx, flags, and │ │ │ more_pending │ │ │ │ │ │ │ │ │ wq_resume(smc_call_ctx) │ │ ├────────────────────────────────►│ │ │ │ │ │ │ │ │ REQUEST_2 COMPLETE │ │ │◄────────────────────────────────┤ │ │ │ │ │ │ With the exception of get_wq_ctx(), the other SMC call wq_resume() can return WQ_SLEEP (these nested rounds of WQ_SLEEP are not shown in the above diagram for the sake of simplicity). Therefore, introduce a new do-while loop to handle multiple WQ_SLEEP return values for the same parent SCM call. Request Completion in the above diagram refers to either a success return value (zero) or error (and not SMC_WAITQ_SLEEP) Also add the interrupt handler that wakes up a sleeping SCM call. Signed-off-by: Guru Das Srinagesh <quic_gurus@quicinc.com> Co-developed-by: Sibi Sankar <quic_sibis@quicinc.com> Signed-off-by: Sibi Sankar <quic_sibis@quicinc.com> Reviewed-by: Guru Das Srinagesh <quic_gurus@quicinc.com> Signed-off-by: Bjorn Andersson <andersson@kernel.org> Link: https://lore.kernel.org/r/20230113161114.22607-3-quic_sibis@quicinc.com
2023-01-14 00:11:14 +08:00
#define QCOM_SMC_WAITQ_FLAG_WAKE_ONE BIT(0)
#define QCOM_SMC_WAITQ_FLAG_WAKE_ALL BIT(1)
static const char * const qcom_scm_convention_names[] = {
[SMC_CONVENTION_UNKNOWN] = "unknown",
[SMC_CONVENTION_ARM_32] = "smc arm 32",
[SMC_CONVENTION_ARM_64] = "smc arm 64",
[SMC_CONVENTION_LEGACY] = "smc legacy",
};
static struct qcom_scm *__scm;
static int qcom_scm_clk_enable(void)
{
int ret;
ret = clk_prepare_enable(__scm->core_clk);
if (ret)
goto bail;
ret = clk_prepare_enable(__scm->iface_clk);
if (ret)
goto disable_core;
ret = clk_prepare_enable(__scm->bus_clk);
if (ret)
goto disable_iface;
return 0;
disable_iface:
clk_disable_unprepare(__scm->iface_clk);
disable_core:
clk_disable_unprepare(__scm->core_clk);
bail:
return ret;
}
static void qcom_scm_clk_disable(void)
{
clk_disable_unprepare(__scm->core_clk);
clk_disable_unprepare(__scm->iface_clk);
clk_disable_unprepare(__scm->bus_clk);
}
static int qcom_scm_bw_enable(void)
{
int ret = 0;
if (!__scm->path)
return 0;
if (IS_ERR(__scm->path))
return -EINVAL;
mutex_lock(&__scm->scm_bw_lock);
if (!__scm->scm_vote_count) {
ret = icc_set_bw(__scm->path, 0, UINT_MAX);
if (ret < 0) {
dev_err(__scm->dev, "failed to set bandwidth request\n");
goto err_bw;
}
}
__scm->scm_vote_count++;
err_bw:
mutex_unlock(&__scm->scm_bw_lock);
return ret;
}
static void qcom_scm_bw_disable(void)
{
if (IS_ERR_OR_NULL(__scm->path))
return;
mutex_lock(&__scm->scm_bw_lock);
if (__scm->scm_vote_count-- == 1)
icc_set_bw(__scm->path, 0, 0);
mutex_unlock(&__scm->scm_bw_lock);
}
firmware: qcom_scm: Reduce locking section for __get_convention() We shouldn't need to hold this spinlock here around the entire SCM call into the firmware and back. Instead, we should be able to query the firmware, potentially in parallel with other CPUs making the same convention detection firmware call, and then grab the lock to update the calling convention detected. The convention doesn't change at runtime so calling into firmware more than once is possibly wasteful but simpler. Besides, this is the slow path, not the fast path where we've already detected the convention used. More importantly, this allows us to add more logic here to workaround the case where the firmware call to check for availability isn't implemented in the firmware at all. In that case we can check the firmware node compatible string and force a calling convention. Note that we remove the 'has_queried' logic that is repeated twice. That could lead to the calling convention being printed multiple times to the kernel logs if the bool is true but __query_convention() is running on multiple CPUs. We also shorten the time where the lock is held, but we keep the lock held around the printk because it doesn't seem hugely important to drop it for that. Cc: Elliot Berman <eberman@codeaurora.org> Cc: Brian Masney <masneyb@onstation.org> Cc: Stephan Gerhold <stephan@gerhold.net> Cc: Jeffrey Hugo <jhugo@codeaurora.org> Cc: Douglas Anderson <dianders@chromium.org> Fixes: 9a434cee773a ("firmware: qcom_scm: Dynamically support SMCCC and legacy conventions") Signed-off-by: Stephen Boyd <swboyd@chromium.org> Link: https://lore.kernel.org/r/20210223214539.1336155-3-swboyd@chromium.org Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
2021-02-24 05:45:35 +08:00
enum qcom_scm_convention qcom_scm_convention = SMC_CONVENTION_UNKNOWN;
static DEFINE_SPINLOCK(scm_query_lock);
firmware: qcom_scm: Reduce locking section for __get_convention() We shouldn't need to hold this spinlock here around the entire SCM call into the firmware and back. Instead, we should be able to query the firmware, potentially in parallel with other CPUs making the same convention detection firmware call, and then grab the lock to update the calling convention detected. The convention doesn't change at runtime so calling into firmware more than once is possibly wasteful but simpler. Besides, this is the slow path, not the fast path where we've already detected the convention used. More importantly, this allows us to add more logic here to workaround the case where the firmware call to check for availability isn't implemented in the firmware at all. In that case we can check the firmware node compatible string and force a calling convention. Note that we remove the 'has_queried' logic that is repeated twice. That could lead to the calling convention being printed multiple times to the kernel logs if the bool is true but __query_convention() is running on multiple CPUs. We also shorten the time where the lock is held, but we keep the lock held around the printk because it doesn't seem hugely important to drop it for that. Cc: Elliot Berman <eberman@codeaurora.org> Cc: Brian Masney <masneyb@onstation.org> Cc: Stephan Gerhold <stephan@gerhold.net> Cc: Jeffrey Hugo <jhugo@codeaurora.org> Cc: Douglas Anderson <dianders@chromium.org> Fixes: 9a434cee773a ("firmware: qcom_scm: Dynamically support SMCCC and legacy conventions") Signed-off-by: Stephen Boyd <swboyd@chromium.org> Link: https://lore.kernel.org/r/20210223214539.1336155-3-swboyd@chromium.org Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
2021-02-24 05:45:35 +08:00
static enum qcom_scm_convention __get_convention(void)
{
unsigned long flags;
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_INFO,
.cmd = QCOM_SCM_INFO_IS_CALL_AVAIL,
.args[0] = SCM_SMC_FNID(QCOM_SCM_SVC_INFO,
QCOM_SCM_INFO_IS_CALL_AVAIL) |
(ARM_SMCCC_OWNER_SIP << ARM_SMCCC_OWNER_SHIFT),
.arginfo = QCOM_SCM_ARGS(1),
.owner = ARM_SMCCC_OWNER_SIP,
};
struct qcom_scm_res res;
firmware: qcom_scm: Reduce locking section for __get_convention() We shouldn't need to hold this spinlock here around the entire SCM call into the firmware and back. Instead, we should be able to query the firmware, potentially in parallel with other CPUs making the same convention detection firmware call, and then grab the lock to update the calling convention detected. The convention doesn't change at runtime so calling into firmware more than once is possibly wasteful but simpler. Besides, this is the slow path, not the fast path where we've already detected the convention used. More importantly, this allows us to add more logic here to workaround the case where the firmware call to check for availability isn't implemented in the firmware at all. In that case we can check the firmware node compatible string and force a calling convention. Note that we remove the 'has_queried' logic that is repeated twice. That could lead to the calling convention being printed multiple times to the kernel logs if the bool is true but __query_convention() is running on multiple CPUs. We also shorten the time where the lock is held, but we keep the lock held around the printk because it doesn't seem hugely important to drop it for that. Cc: Elliot Berman <eberman@codeaurora.org> Cc: Brian Masney <masneyb@onstation.org> Cc: Stephan Gerhold <stephan@gerhold.net> Cc: Jeffrey Hugo <jhugo@codeaurora.org> Cc: Douglas Anderson <dianders@chromium.org> Fixes: 9a434cee773a ("firmware: qcom_scm: Dynamically support SMCCC and legacy conventions") Signed-off-by: Stephen Boyd <swboyd@chromium.org> Link: https://lore.kernel.org/r/20210223214539.1336155-3-swboyd@chromium.org Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
2021-02-24 05:45:35 +08:00
enum qcom_scm_convention probed_convention;
int ret;
firmware: qcom_scm: Workaround lack of "is available" call on SC7180 Some SC7180 firmwares don't implement the QCOM_SCM_INFO_IS_CALL_AVAIL API, so we can't probe the calling convention. We detect the legacy calling convention on these firmwares, because the availability call always fails and legacy is the fallback. This leads to problems where the rmtfs driver fails to probe, because it tries to assign memory with a bad calling convention, which then leads to modem failing to load and all networking, even wifi, to fail. Ouch! Let's force the calling convention to be what it always is on this SoC, i.e. arm64. Of course, the calling convention is not the same thing as implementing the QCOM_SCM_INFO_IS_CALL_AVAIL API. The absence of the "is this call available" API from the firmware means that any call to __qcom_scm_is_call_available() fails. This is OK for now though because none of the calls that are checked for existence are implemented on firmware running on sc7180. If such a call needs to be checked for existence in the future, we presume that firmware will implement this API and then things will "just work". Cc: Elliot Berman <eberman@codeaurora.org> Cc: Brian Masney <masneyb@onstation.org> Cc: Stephan Gerhold <stephan@gerhold.net> Cc: Jeffrey Hugo <jhugo@codeaurora.org> Cc: Douglas Anderson <dianders@chromium.org> Fixes: 9a434cee773a ("firmware: qcom_scm: Dynamically support SMCCC and legacy conventions") Signed-off-by: Stephen Boyd <swboyd@chromium.org> Link: https://lore.kernel.org/r/20210223214539.1336155-4-swboyd@chromium.org Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
2021-02-24 05:45:36 +08:00
bool forced = false;
firmware: qcom_scm: Reduce locking section for __get_convention() We shouldn't need to hold this spinlock here around the entire SCM call into the firmware and back. Instead, we should be able to query the firmware, potentially in parallel with other CPUs making the same convention detection firmware call, and then grab the lock to update the calling convention detected. The convention doesn't change at runtime so calling into firmware more than once is possibly wasteful but simpler. Besides, this is the slow path, not the fast path where we've already detected the convention used. More importantly, this allows us to add more logic here to workaround the case where the firmware call to check for availability isn't implemented in the firmware at all. In that case we can check the firmware node compatible string and force a calling convention. Note that we remove the 'has_queried' logic that is repeated twice. That could lead to the calling convention being printed multiple times to the kernel logs if the bool is true but __query_convention() is running on multiple CPUs. We also shorten the time where the lock is held, but we keep the lock held around the printk because it doesn't seem hugely important to drop it for that. Cc: Elliot Berman <eberman@codeaurora.org> Cc: Brian Masney <masneyb@onstation.org> Cc: Stephan Gerhold <stephan@gerhold.net> Cc: Jeffrey Hugo <jhugo@codeaurora.org> Cc: Douglas Anderson <dianders@chromium.org> Fixes: 9a434cee773a ("firmware: qcom_scm: Dynamically support SMCCC and legacy conventions") Signed-off-by: Stephen Boyd <swboyd@chromium.org> Link: https://lore.kernel.org/r/20210223214539.1336155-3-swboyd@chromium.org Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
2021-02-24 05:45:35 +08:00
if (likely(qcom_scm_convention != SMC_CONVENTION_UNKNOWN))
return qcom_scm_convention;
firmware: qcom_scm: Reduce locking section for __get_convention() We shouldn't need to hold this spinlock here around the entire SCM call into the firmware and back. Instead, we should be able to query the firmware, potentially in parallel with other CPUs making the same convention detection firmware call, and then grab the lock to update the calling convention detected. The convention doesn't change at runtime so calling into firmware more than once is possibly wasteful but simpler. Besides, this is the slow path, not the fast path where we've already detected the convention used. More importantly, this allows us to add more logic here to workaround the case where the firmware call to check for availability isn't implemented in the firmware at all. In that case we can check the firmware node compatible string and force a calling convention. Note that we remove the 'has_queried' logic that is repeated twice. That could lead to the calling convention being printed multiple times to the kernel logs if the bool is true but __query_convention() is running on multiple CPUs. We also shorten the time where the lock is held, but we keep the lock held around the printk because it doesn't seem hugely important to drop it for that. Cc: Elliot Berman <eberman@codeaurora.org> Cc: Brian Masney <masneyb@onstation.org> Cc: Stephan Gerhold <stephan@gerhold.net> Cc: Jeffrey Hugo <jhugo@codeaurora.org> Cc: Douglas Anderson <dianders@chromium.org> Fixes: 9a434cee773a ("firmware: qcom_scm: Dynamically support SMCCC and legacy conventions") Signed-off-by: Stephen Boyd <swboyd@chromium.org> Link: https://lore.kernel.org/r/20210223214539.1336155-3-swboyd@chromium.org Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
2021-02-24 05:45:35 +08:00
/*
* Device isn't required as there is only one argument - no device
* needed to dma_map_single to secure world
*/
probed_convention = SMC_CONVENTION_ARM_64;
ret = __scm_smc_call(NULL, &desc, probed_convention, &res, true);
if (!ret && res.result[0] == 1)
firmware: qcom_scm: Reduce locking section for __get_convention() We shouldn't need to hold this spinlock here around the entire SCM call into the firmware and back. Instead, we should be able to query the firmware, potentially in parallel with other CPUs making the same convention detection firmware call, and then grab the lock to update the calling convention detected. The convention doesn't change at runtime so calling into firmware more than once is possibly wasteful but simpler. Besides, this is the slow path, not the fast path where we've already detected the convention used. More importantly, this allows us to add more logic here to workaround the case where the firmware call to check for availability isn't implemented in the firmware at all. In that case we can check the firmware node compatible string and force a calling convention. Note that we remove the 'has_queried' logic that is repeated twice. That could lead to the calling convention being printed multiple times to the kernel logs if the bool is true but __query_convention() is running on multiple CPUs. We also shorten the time where the lock is held, but we keep the lock held around the printk because it doesn't seem hugely important to drop it for that. Cc: Elliot Berman <eberman@codeaurora.org> Cc: Brian Masney <masneyb@onstation.org> Cc: Stephan Gerhold <stephan@gerhold.net> Cc: Jeffrey Hugo <jhugo@codeaurora.org> Cc: Douglas Anderson <dianders@chromium.org> Fixes: 9a434cee773a ("firmware: qcom_scm: Dynamically support SMCCC and legacy conventions") Signed-off-by: Stephen Boyd <swboyd@chromium.org> Link: https://lore.kernel.org/r/20210223214539.1336155-3-swboyd@chromium.org Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
2021-02-24 05:45:35 +08:00
goto found;
firmware: qcom_scm: Workaround lack of "is available" call on SC7180 Some SC7180 firmwares don't implement the QCOM_SCM_INFO_IS_CALL_AVAIL API, so we can't probe the calling convention. We detect the legacy calling convention on these firmwares, because the availability call always fails and legacy is the fallback. This leads to problems where the rmtfs driver fails to probe, because it tries to assign memory with a bad calling convention, which then leads to modem failing to load and all networking, even wifi, to fail. Ouch! Let's force the calling convention to be what it always is on this SoC, i.e. arm64. Of course, the calling convention is not the same thing as implementing the QCOM_SCM_INFO_IS_CALL_AVAIL API. The absence of the "is this call available" API from the firmware means that any call to __qcom_scm_is_call_available() fails. This is OK for now though because none of the calls that are checked for existence are implemented on firmware running on sc7180. If such a call needs to be checked for existence in the future, we presume that firmware will implement this API and then things will "just work". Cc: Elliot Berman <eberman@codeaurora.org> Cc: Brian Masney <masneyb@onstation.org> Cc: Stephan Gerhold <stephan@gerhold.net> Cc: Jeffrey Hugo <jhugo@codeaurora.org> Cc: Douglas Anderson <dianders@chromium.org> Fixes: 9a434cee773a ("firmware: qcom_scm: Dynamically support SMCCC and legacy conventions") Signed-off-by: Stephen Boyd <swboyd@chromium.org> Link: https://lore.kernel.org/r/20210223214539.1336155-4-swboyd@chromium.org Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
2021-02-24 05:45:36 +08:00
/*
* Some SC7180 firmwares didn't implement the
* QCOM_SCM_INFO_IS_CALL_AVAIL call, so we fallback to forcing ARM_64
* calling conventions on these firmwares. Luckily we don't make any
* early calls into the firmware on these SoCs so the device pointer
* will be valid here to check if the compatible matches.
*/
if (of_device_is_compatible(__scm ? __scm->dev->of_node : NULL, "qcom,scm-sc7180")) {
forced = true;
goto found;
}
firmware: qcom_scm: Reduce locking section for __get_convention() We shouldn't need to hold this spinlock here around the entire SCM call into the firmware and back. Instead, we should be able to query the firmware, potentially in parallel with other CPUs making the same convention detection firmware call, and then grab the lock to update the calling convention detected. The convention doesn't change at runtime so calling into firmware more than once is possibly wasteful but simpler. Besides, this is the slow path, not the fast path where we've already detected the convention used. More importantly, this allows us to add more logic here to workaround the case where the firmware call to check for availability isn't implemented in the firmware at all. In that case we can check the firmware node compatible string and force a calling convention. Note that we remove the 'has_queried' logic that is repeated twice. That could lead to the calling convention being printed multiple times to the kernel logs if the bool is true but __query_convention() is running on multiple CPUs. We also shorten the time where the lock is held, but we keep the lock held around the printk because it doesn't seem hugely important to drop it for that. Cc: Elliot Berman <eberman@codeaurora.org> Cc: Brian Masney <masneyb@onstation.org> Cc: Stephan Gerhold <stephan@gerhold.net> Cc: Jeffrey Hugo <jhugo@codeaurora.org> Cc: Douglas Anderson <dianders@chromium.org> Fixes: 9a434cee773a ("firmware: qcom_scm: Dynamically support SMCCC and legacy conventions") Signed-off-by: Stephen Boyd <swboyd@chromium.org> Link: https://lore.kernel.org/r/20210223214539.1336155-3-swboyd@chromium.org Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
2021-02-24 05:45:35 +08:00
probed_convention = SMC_CONVENTION_ARM_32;
ret = __scm_smc_call(NULL, &desc, probed_convention, &res, true);
if (!ret && res.result[0] == 1)
firmware: qcom_scm: Reduce locking section for __get_convention() We shouldn't need to hold this spinlock here around the entire SCM call into the firmware and back. Instead, we should be able to query the firmware, potentially in parallel with other CPUs making the same convention detection firmware call, and then grab the lock to update the calling convention detected. The convention doesn't change at runtime so calling into firmware more than once is possibly wasteful but simpler. Besides, this is the slow path, not the fast path where we've already detected the convention used. More importantly, this allows us to add more logic here to workaround the case where the firmware call to check for availability isn't implemented in the firmware at all. In that case we can check the firmware node compatible string and force a calling convention. Note that we remove the 'has_queried' logic that is repeated twice. That could lead to the calling convention being printed multiple times to the kernel logs if the bool is true but __query_convention() is running on multiple CPUs. We also shorten the time where the lock is held, but we keep the lock held around the printk because it doesn't seem hugely important to drop it for that. Cc: Elliot Berman <eberman@codeaurora.org> Cc: Brian Masney <masneyb@onstation.org> Cc: Stephan Gerhold <stephan@gerhold.net> Cc: Jeffrey Hugo <jhugo@codeaurora.org> Cc: Douglas Anderson <dianders@chromium.org> Fixes: 9a434cee773a ("firmware: qcom_scm: Dynamically support SMCCC and legacy conventions") Signed-off-by: Stephen Boyd <swboyd@chromium.org> Link: https://lore.kernel.org/r/20210223214539.1336155-3-swboyd@chromium.org Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
2021-02-24 05:45:35 +08:00
goto found;
probed_convention = SMC_CONVENTION_LEGACY;
found:
spin_lock_irqsave(&scm_query_lock, flags);
if (probed_convention != qcom_scm_convention) {
qcom_scm_convention = probed_convention;
firmware: qcom_scm: Workaround lack of "is available" call on SC7180 Some SC7180 firmwares don't implement the QCOM_SCM_INFO_IS_CALL_AVAIL API, so we can't probe the calling convention. We detect the legacy calling convention on these firmwares, because the availability call always fails and legacy is the fallback. This leads to problems where the rmtfs driver fails to probe, because it tries to assign memory with a bad calling convention, which then leads to modem failing to load and all networking, even wifi, to fail. Ouch! Let's force the calling convention to be what it always is on this SoC, i.e. arm64. Of course, the calling convention is not the same thing as implementing the QCOM_SCM_INFO_IS_CALL_AVAIL API. The absence of the "is this call available" API from the firmware means that any call to __qcom_scm_is_call_available() fails. This is OK for now though because none of the calls that are checked for existence are implemented on firmware running on sc7180. If such a call needs to be checked for existence in the future, we presume that firmware will implement this API and then things will "just work". Cc: Elliot Berman <eberman@codeaurora.org> Cc: Brian Masney <masneyb@onstation.org> Cc: Stephan Gerhold <stephan@gerhold.net> Cc: Jeffrey Hugo <jhugo@codeaurora.org> Cc: Douglas Anderson <dianders@chromium.org> Fixes: 9a434cee773a ("firmware: qcom_scm: Dynamically support SMCCC and legacy conventions") Signed-off-by: Stephen Boyd <swboyd@chromium.org> Link: https://lore.kernel.org/r/20210223214539.1336155-4-swboyd@chromium.org Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
2021-02-24 05:45:36 +08:00
pr_info("qcom_scm: convention: %s%s\n",
qcom_scm_convention_names[qcom_scm_convention],
forced ? " (forced)" : "");
firmware: qcom_scm: Reduce locking section for __get_convention() We shouldn't need to hold this spinlock here around the entire SCM call into the firmware and back. Instead, we should be able to query the firmware, potentially in parallel with other CPUs making the same convention detection firmware call, and then grab the lock to update the calling convention detected. The convention doesn't change at runtime so calling into firmware more than once is possibly wasteful but simpler. Besides, this is the slow path, not the fast path where we've already detected the convention used. More importantly, this allows us to add more logic here to workaround the case where the firmware call to check for availability isn't implemented in the firmware at all. In that case we can check the firmware node compatible string and force a calling convention. Note that we remove the 'has_queried' logic that is repeated twice. That could lead to the calling convention being printed multiple times to the kernel logs if the bool is true but __query_convention() is running on multiple CPUs. We also shorten the time where the lock is held, but we keep the lock held around the printk because it doesn't seem hugely important to drop it for that. Cc: Elliot Berman <eberman@codeaurora.org> Cc: Brian Masney <masneyb@onstation.org> Cc: Stephan Gerhold <stephan@gerhold.net> Cc: Jeffrey Hugo <jhugo@codeaurora.org> Cc: Douglas Anderson <dianders@chromium.org> Fixes: 9a434cee773a ("firmware: qcom_scm: Dynamically support SMCCC and legacy conventions") Signed-off-by: Stephen Boyd <swboyd@chromium.org> Link: https://lore.kernel.org/r/20210223214539.1336155-3-swboyd@chromium.org Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
2021-02-24 05:45:35 +08:00
}
spin_unlock_irqrestore(&scm_query_lock, flags);
return qcom_scm_convention;
}
/**
* qcom_scm_call() - Invoke a syscall in the secure world
* @dev: device
* @desc: Descriptor structure containing arguments and return values
* @res: Structure containing results from SMC/HVC call
*
* Sends a command to the SCM and waits for the command to finish processing.
* This should *only* be called in pre-emptible context.
*/
static int qcom_scm_call(struct device *dev, const struct qcom_scm_desc *desc,
struct qcom_scm_res *res)
{
might_sleep();
switch (__get_convention()) {
case SMC_CONVENTION_ARM_32:
case SMC_CONVENTION_ARM_64:
return scm_smc_call(dev, desc, res, false);
case SMC_CONVENTION_LEGACY:
return scm_legacy_call(dev, desc, res);
default:
pr_err("Unknown current SCM calling convention.\n");
return -EINVAL;
}
}
/**
* qcom_scm_call_atomic() - atomic variation of qcom_scm_call()
* @dev: device
* @desc: Descriptor structure containing arguments and return values
* @res: Structure containing results from SMC/HVC call
*
* Sends a command to the SCM and waits for the command to finish processing.
* This can be called in atomic context.
*/
static int qcom_scm_call_atomic(struct device *dev,
const struct qcom_scm_desc *desc,
struct qcom_scm_res *res)
{
switch (__get_convention()) {
case SMC_CONVENTION_ARM_32:
case SMC_CONVENTION_ARM_64:
return scm_smc_call(dev, desc, res, true);
case SMC_CONVENTION_LEGACY:
return scm_legacy_call_atomic(dev, desc, res);
default:
pr_err("Unknown current SCM calling convention.\n");
return -EINVAL;
}
}
static bool __qcom_scm_is_call_available(struct device *dev, u32 svc_id,
u32 cmd_id)
{
int ret;
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_INFO,
.cmd = QCOM_SCM_INFO_IS_CALL_AVAIL,
.owner = ARM_SMCCC_OWNER_SIP,
};
struct qcom_scm_res res;
desc.arginfo = QCOM_SCM_ARGS(1);
switch (__get_convention()) {
case SMC_CONVENTION_ARM_32:
case SMC_CONVENTION_ARM_64:
desc.args[0] = SCM_SMC_FNID(svc_id, cmd_id) |
(ARM_SMCCC_OWNER_SIP << ARM_SMCCC_OWNER_SHIFT);
break;
case SMC_CONVENTION_LEGACY:
desc.args[0] = SCM_LEGACY_FNID(svc_id, cmd_id);
break;
default:
pr_err("Unknown SMC convention being used\n");
return false;
}
ret = qcom_scm_call(dev, &desc, &res);
return ret ? false : !!res.result[0];
}
static int qcom_scm_set_boot_addr(void *entry, const u8 *cpu_bits)
{
int cpu;
unsigned int flags = 0;
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_BOOT,
.cmd = QCOM_SCM_BOOT_SET_ADDR,
.arginfo = QCOM_SCM_ARGS(2),
.owner = ARM_SMCCC_OWNER_SIP,
};
for_each_present_cpu(cpu) {
if (cpu >= QCOM_SCM_BOOT_MAX_CPUS)
return -EINVAL;
flags |= cpu_bits[cpu];
}
desc.args[0] = flags;
desc.args[1] = virt_to_phys(entry);
return qcom_scm_call_atomic(__scm ? __scm->dev : NULL, &desc, NULL);
}
static int qcom_scm_set_boot_addr_mc(void *entry, unsigned int flags)
{
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_BOOT,
.cmd = QCOM_SCM_BOOT_SET_ADDR_MC,
.owner = ARM_SMCCC_OWNER_SIP,
.arginfo = QCOM_SCM_ARGS(6),
.args = {
virt_to_phys(entry),
/* Apply to all CPUs in all affinity levels */
~0ULL, ~0ULL, ~0ULL, ~0ULL,
flags,
},
};
/* Need a device for DMA of the additional arguments */
if (!__scm || __get_convention() == SMC_CONVENTION_LEGACY)
return -EOPNOTSUPP;
return qcom_scm_call(__scm->dev, &desc, NULL);
}
/**
* qcom_scm_set_warm_boot_addr() - Set the warm boot address for all cpus
* @entry: Entry point function for the cpus
*
* Set the Linux entry point for the SCM to transfer control to when coming
* out of a power down. CPU power down may be executed on cpuidle or hotplug.
*/
int qcom_scm_set_warm_boot_addr(void *entry)
{
if (qcom_scm_set_boot_addr_mc(entry, QCOM_SCM_BOOT_MC_FLAG_WARMBOOT))
/* Fallback to old SCM call */
return qcom_scm_set_boot_addr(entry, qcom_scm_cpu_warm_bits);
return 0;
}
EXPORT_SYMBOL(qcom_scm_set_warm_boot_addr);
/**
* qcom_scm_set_cold_boot_addr() - Set the cold boot address for all cpus
* @entry: Entry point function for the cpus
*/
int qcom_scm_set_cold_boot_addr(void *entry)
{
if (qcom_scm_set_boot_addr_mc(entry, QCOM_SCM_BOOT_MC_FLAG_COLDBOOT))
/* Fallback to old SCM call */
return qcom_scm_set_boot_addr(entry, qcom_scm_cpu_cold_bits);
return 0;
}
EXPORT_SYMBOL(qcom_scm_set_cold_boot_addr);
/**
* qcom_scm_cpu_power_down() - Power down the cpu
* @flags: Flags to flush cache
*
* This is an end point to power down cpu. If there was a pending interrupt,
* the control would return from this function, otherwise, the cpu jumps to the
* warm boot entry point set for this cpu upon reset.
*/
void qcom_scm_cpu_power_down(u32 flags)
{
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_BOOT,
.cmd = QCOM_SCM_BOOT_TERMINATE_PC,
.args[0] = flags & QCOM_SCM_FLUSH_FLAG_MASK,
.arginfo = QCOM_SCM_ARGS(1),
.owner = ARM_SMCCC_OWNER_SIP,
};
qcom_scm_call_atomic(__scm ? __scm->dev : NULL, &desc, NULL);
}
EXPORT_SYMBOL(qcom_scm_cpu_power_down);
int qcom_scm_set_remote_state(u32 state, u32 id)
{
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_BOOT,
.cmd = QCOM_SCM_BOOT_SET_REMOTE_STATE,
.arginfo = QCOM_SCM_ARGS(2),
.args[0] = state,
.args[1] = id,
.owner = ARM_SMCCC_OWNER_SIP,
};
struct qcom_scm_res res;
int ret;
ret = qcom_scm_call(__scm->dev, &desc, &res);
return ret ? : res.result[0];
}
EXPORT_SYMBOL(qcom_scm_set_remote_state);
static int __qcom_scm_set_dload_mode(struct device *dev, bool enable)
{
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_BOOT,
.cmd = QCOM_SCM_BOOT_SET_DLOAD_MODE,
.arginfo = QCOM_SCM_ARGS(2),
.args[0] = QCOM_SCM_BOOT_SET_DLOAD_MODE,
.owner = ARM_SMCCC_OWNER_SIP,
};
desc.args[1] = enable ? QCOM_SCM_BOOT_SET_DLOAD_MODE : 0;
return qcom_scm_call_atomic(__scm->dev, &desc, NULL);
}
static void qcom_scm_set_download_mode(bool enable)
{
bool avail;
int ret = 0;
avail = __qcom_scm_is_call_available(__scm->dev,
QCOM_SCM_SVC_BOOT,
QCOM_SCM_BOOT_SET_DLOAD_MODE);
if (avail) {
ret = __qcom_scm_set_dload_mode(__scm->dev, enable);
} else if (__scm->dload_mode_addr) {
ret = qcom_scm_io_writel(__scm->dload_mode_addr,
enable ? QCOM_SCM_BOOT_SET_DLOAD_MODE : 0);
} else {
dev_err(__scm->dev,
"No available mechanism for setting download mode\n");
}
if (ret)
dev_err(__scm->dev, "failed to set download mode: %d\n", ret);
}
/**
* qcom_scm_pas_init_image() - Initialize peripheral authentication service
* state machine for a given peripheral, using the
* metadata
* @peripheral: peripheral id
* @metadata: pointer to memory containing ELF header, program header table
* and optional blob of data used for authenticating the metadata
* and the rest of the firmware
* @size: size of the metadata
* @ctx: optional metadata context
*
* Return: 0 on success.
*
* Upon successful return, the PAS metadata context (@ctx) will be used to
* track the metadata allocation, this needs to be released by invoking
* qcom_scm_pas_metadata_release() by the caller.
*/
int qcom_scm_pas_init_image(u32 peripheral, const void *metadata, size_t size,
struct qcom_scm_pas_metadata *ctx)
{
dma_addr_t mdata_phys;
void *mdata_buf;
int ret;
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_PIL,
.cmd = QCOM_SCM_PIL_PAS_INIT_IMAGE,
.arginfo = QCOM_SCM_ARGS(2, QCOM_SCM_VAL, QCOM_SCM_RW),
.args[0] = peripheral,
.owner = ARM_SMCCC_OWNER_SIP,
};
struct qcom_scm_res res;
/*
* During the scm call memory protection will be enabled for the meta
* data blob, so make sure it's physically contiguous, 4K aligned and
* non-cachable to avoid XPU violations.
*/
mdata_buf = dma_alloc_coherent(__scm->dev, size, &mdata_phys,
GFP_KERNEL);
if (!mdata_buf) {
dev_err(__scm->dev, "Allocation of metadata buffer failed.\n");
return -ENOMEM;
}
memcpy(mdata_buf, metadata, size);
ret = qcom_scm_clk_enable();
if (ret)
goto out;
ret = qcom_scm_bw_enable();
if (ret)
return ret;
desc.args[1] = mdata_phys;
ret = qcom_scm_call(__scm->dev, &desc, &res);
qcom_scm_bw_disable();
qcom_scm_clk_disable();
out:
if (ret < 0 || !ctx) {
dma_free_coherent(__scm->dev, size, mdata_buf, mdata_phys);
} else if (ctx) {
ctx->ptr = mdata_buf;
ctx->phys = mdata_phys;
ctx->size = size;
}
return ret ? : res.result[0];
}
EXPORT_SYMBOL(qcom_scm_pas_init_image);
/**
* qcom_scm_pas_metadata_release() - release metadata context
* @ctx: metadata context
*/
void qcom_scm_pas_metadata_release(struct qcom_scm_pas_metadata *ctx)
{
if (!ctx->ptr)
return;
dma_free_coherent(__scm->dev, ctx->size, ctx->ptr, ctx->phys);
ctx->ptr = NULL;
ctx->phys = 0;
ctx->size = 0;
}
EXPORT_SYMBOL(qcom_scm_pas_metadata_release);
/**
* qcom_scm_pas_mem_setup() - Prepare the memory related to a given peripheral
* for firmware loading
* @peripheral: peripheral id
* @addr: start address of memory area to prepare
* @size: size of the memory area to prepare
*
* Returns 0 on success.
*/
int qcom_scm_pas_mem_setup(u32 peripheral, phys_addr_t addr, phys_addr_t size)
{
int ret;
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_PIL,
.cmd = QCOM_SCM_PIL_PAS_MEM_SETUP,
.arginfo = QCOM_SCM_ARGS(3),
.args[0] = peripheral,
.args[1] = addr,
.args[2] = size,
.owner = ARM_SMCCC_OWNER_SIP,
};
struct qcom_scm_res res;
ret = qcom_scm_clk_enable();
if (ret)
return ret;
ret = qcom_scm_bw_enable();
if (ret)
return ret;
ret = qcom_scm_call(__scm->dev, &desc, &res);
qcom_scm_bw_disable();
qcom_scm_clk_disable();
return ret ? : res.result[0];
}
EXPORT_SYMBOL(qcom_scm_pas_mem_setup);
/**
* qcom_scm_pas_auth_and_reset() - Authenticate the given peripheral firmware
* and reset the remote processor
* @peripheral: peripheral id
*
* Return 0 on success.
*/
int qcom_scm_pas_auth_and_reset(u32 peripheral)
{
int ret;
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_PIL,
.cmd = QCOM_SCM_PIL_PAS_AUTH_AND_RESET,
.arginfo = QCOM_SCM_ARGS(1),
.args[0] = peripheral,
.owner = ARM_SMCCC_OWNER_SIP,
};
struct qcom_scm_res res;
ret = qcom_scm_clk_enable();
if (ret)
return ret;
ret = qcom_scm_bw_enable();
if (ret)
return ret;
ret = qcom_scm_call(__scm->dev, &desc, &res);
qcom_scm_bw_disable();
qcom_scm_clk_disable();
return ret ? : res.result[0];
}
EXPORT_SYMBOL(qcom_scm_pas_auth_and_reset);
/**
* qcom_scm_pas_shutdown() - Shut down the remote processor
* @peripheral: peripheral id
*
* Returns 0 on success.
*/
int qcom_scm_pas_shutdown(u32 peripheral)
{
int ret;
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_PIL,
.cmd = QCOM_SCM_PIL_PAS_SHUTDOWN,
.arginfo = QCOM_SCM_ARGS(1),
.args[0] = peripheral,
.owner = ARM_SMCCC_OWNER_SIP,
};
struct qcom_scm_res res;
ret = qcom_scm_clk_enable();
if (ret)
return ret;
ret = qcom_scm_bw_enable();
if (ret)
return ret;
ret = qcom_scm_call(__scm->dev, &desc, &res);
qcom_scm_bw_disable();
qcom_scm_clk_disable();
return ret ? : res.result[0];
}
EXPORT_SYMBOL(qcom_scm_pas_shutdown);
/**
* qcom_scm_pas_supported() - Check if the peripheral authentication service is
* available for the given peripherial
* @peripheral: peripheral id
*
* Returns true if PAS is supported for this peripheral, otherwise false.
*/
bool qcom_scm_pas_supported(u32 peripheral)
{
int ret;
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_PIL,
.cmd = QCOM_SCM_PIL_PAS_IS_SUPPORTED,
.arginfo = QCOM_SCM_ARGS(1),
.args[0] = peripheral,
.owner = ARM_SMCCC_OWNER_SIP,
};
struct qcom_scm_res res;
if (!__qcom_scm_is_call_available(__scm->dev, QCOM_SCM_SVC_PIL,
QCOM_SCM_PIL_PAS_IS_SUPPORTED))
return false;
ret = qcom_scm_call(__scm->dev, &desc, &res);
return ret ? false : !!res.result[0];
}
EXPORT_SYMBOL(qcom_scm_pas_supported);
static int __qcom_scm_pas_mss_reset(struct device *dev, bool reset)
{
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_PIL,
.cmd = QCOM_SCM_PIL_PAS_MSS_RESET,
.arginfo = QCOM_SCM_ARGS(2),
.args[0] = reset,
.args[1] = 0,
.owner = ARM_SMCCC_OWNER_SIP,
};
struct qcom_scm_res res;
int ret;
ret = qcom_scm_call(__scm->dev, &desc, &res);
return ret ? : res.result[0];
}
static int qcom_scm_pas_reset_assert(struct reset_controller_dev *rcdev,
unsigned long idx)
{
if (idx != 0)
return -EINVAL;
return __qcom_scm_pas_mss_reset(__scm->dev, 1);
}
static int qcom_scm_pas_reset_deassert(struct reset_controller_dev *rcdev,
unsigned long idx)
{
if (idx != 0)
return -EINVAL;
return __qcom_scm_pas_mss_reset(__scm->dev, 0);
}
static const struct reset_control_ops qcom_scm_pas_reset_ops = {
.assert = qcom_scm_pas_reset_assert,
.deassert = qcom_scm_pas_reset_deassert,
};
int qcom_scm_io_readl(phys_addr_t addr, unsigned int *val)
{
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_IO,
.cmd = QCOM_SCM_IO_READ,
.arginfo = QCOM_SCM_ARGS(1),
.args[0] = addr,
.owner = ARM_SMCCC_OWNER_SIP,
};
struct qcom_scm_res res;
int ret;
ret = qcom_scm_call_atomic(__scm->dev, &desc, &res);
if (ret >= 0)
*val = res.result[0];
return ret < 0 ? ret : 0;
}
EXPORT_SYMBOL(qcom_scm_io_readl);
int qcom_scm_io_writel(phys_addr_t addr, unsigned int val)
{
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_IO,
.cmd = QCOM_SCM_IO_WRITE,
.arginfo = QCOM_SCM_ARGS(2),
.args[0] = addr,
.args[1] = val,
.owner = ARM_SMCCC_OWNER_SIP,
};
return qcom_scm_call_atomic(__scm->dev, &desc, NULL);
}
EXPORT_SYMBOL(qcom_scm_io_writel);
/**
* qcom_scm_restore_sec_cfg_available() - Check if secure environment
* supports restore security config interface.
*
* Return true if restore-cfg interface is supported, false if not.
*/
bool qcom_scm_restore_sec_cfg_available(void)
{
return __qcom_scm_is_call_available(__scm->dev, QCOM_SCM_SVC_MP,
QCOM_SCM_MP_RESTORE_SEC_CFG);
}
EXPORT_SYMBOL(qcom_scm_restore_sec_cfg_available);
int qcom_scm_restore_sec_cfg(u32 device_id, u32 spare)
{
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_MP,
.cmd = QCOM_SCM_MP_RESTORE_SEC_CFG,
.arginfo = QCOM_SCM_ARGS(2),
.args[0] = device_id,
.args[1] = spare,
.owner = ARM_SMCCC_OWNER_SIP,
};
struct qcom_scm_res res;
int ret;
ret = qcom_scm_call(__scm->dev, &desc, &res);
return ret ? : res.result[0];
}
EXPORT_SYMBOL(qcom_scm_restore_sec_cfg);
int qcom_scm_iommu_secure_ptbl_size(u32 spare, size_t *size)
{
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_MP,
.cmd = QCOM_SCM_MP_IOMMU_SECURE_PTBL_SIZE,
.arginfo = QCOM_SCM_ARGS(1),
.args[0] = spare,
.owner = ARM_SMCCC_OWNER_SIP,
};
struct qcom_scm_res res;
int ret;
ret = qcom_scm_call(__scm->dev, &desc, &res);
if (size)
*size = res.result[0];
return ret ? : res.result[1];
}
EXPORT_SYMBOL(qcom_scm_iommu_secure_ptbl_size);
int qcom_scm_iommu_secure_ptbl_init(u64 addr, u32 size, u32 spare)
{
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_MP,
.cmd = QCOM_SCM_MP_IOMMU_SECURE_PTBL_INIT,
.arginfo = QCOM_SCM_ARGS(3, QCOM_SCM_RW, QCOM_SCM_VAL,
QCOM_SCM_VAL),
.args[0] = addr,
.args[1] = size,
.args[2] = spare,
.owner = ARM_SMCCC_OWNER_SIP,
};
int ret;
ret = qcom_scm_call(__scm->dev, &desc, NULL);
/* the pg table has been initialized already, ignore the error */
if (ret == -EPERM)
ret = 0;
return ret;
}
EXPORT_SYMBOL(qcom_scm_iommu_secure_ptbl_init);
int qcom_scm_iommu_set_cp_pool_size(u32 spare, u32 size)
{
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_MP,
.cmd = QCOM_SCM_MP_IOMMU_SET_CP_POOL_SIZE,
.arginfo = QCOM_SCM_ARGS(2),
.args[0] = size,
.args[1] = spare,
.owner = ARM_SMCCC_OWNER_SIP,
};
return qcom_scm_call(__scm->dev, &desc, NULL);
}
EXPORT_SYMBOL(qcom_scm_iommu_set_cp_pool_size);
int qcom_scm_mem_protect_video_var(u32 cp_start, u32 cp_size,
u32 cp_nonpixel_start,
u32 cp_nonpixel_size)
{
int ret;
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_MP,
.cmd = QCOM_SCM_MP_VIDEO_VAR,
.arginfo = QCOM_SCM_ARGS(4, QCOM_SCM_VAL, QCOM_SCM_VAL,
QCOM_SCM_VAL, QCOM_SCM_VAL),
.args[0] = cp_start,
.args[1] = cp_size,
.args[2] = cp_nonpixel_start,
.args[3] = cp_nonpixel_size,
.owner = ARM_SMCCC_OWNER_SIP,
};
struct qcom_scm_res res;
ret = qcom_scm_call(__scm->dev, &desc, &res);
return ret ? : res.result[0];
}
EXPORT_SYMBOL(qcom_scm_mem_protect_video_var);
static int __qcom_scm_assign_mem(struct device *dev, phys_addr_t mem_region,
size_t mem_sz, phys_addr_t src, size_t src_sz,
phys_addr_t dest, size_t dest_sz)
{
int ret;
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_MP,
.cmd = QCOM_SCM_MP_ASSIGN,
.arginfo = QCOM_SCM_ARGS(7, QCOM_SCM_RO, QCOM_SCM_VAL,
QCOM_SCM_RO, QCOM_SCM_VAL, QCOM_SCM_RO,
QCOM_SCM_VAL, QCOM_SCM_VAL),
.args[0] = mem_region,
.args[1] = mem_sz,
.args[2] = src,
.args[3] = src_sz,
.args[4] = dest,
.args[5] = dest_sz,
.args[6] = 0,
.owner = ARM_SMCCC_OWNER_SIP,
};
struct qcom_scm_res res;
ret = qcom_scm_call(dev, &desc, &res);
return ret ? : res.result[0];
}
/**
* qcom_scm_assign_mem() - Make a secure call to reassign memory ownership
* @mem_addr: mem region whose ownership need to be reassigned
* @mem_sz: size of the region.
* @srcvm: vmid for current set of owners, each set bit in
* flag indicate a unique owner
* @newvm: array having new owners and corresponding permission
* flags
* @dest_cnt: number of owners in next set.
*
* Return negative errno on failure or 0 on success with @srcvm updated.
*/
int qcom_scm_assign_mem(phys_addr_t mem_addr, size_t mem_sz,
u64 *srcvm,
const struct qcom_scm_vmperm *newvm,
unsigned int dest_cnt)
{
struct qcom_scm_current_perm_info *destvm;
struct qcom_scm_mem_map_info *mem_to_map;
phys_addr_t mem_to_map_phys;
phys_addr_t dest_phys;
dma_addr_t ptr_phys;
size_t mem_to_map_sz;
size_t dest_sz;
size_t src_sz;
size_t ptr_sz;
int next_vm;
__le32 *src;
void *ptr;
int ret, i, b;
u64 srcvm_bits = *srcvm;
src_sz = hweight64(srcvm_bits) * sizeof(*src);
mem_to_map_sz = sizeof(*mem_to_map);
dest_sz = dest_cnt * sizeof(*destvm);
ptr_sz = ALIGN(src_sz, SZ_64) + ALIGN(mem_to_map_sz, SZ_64) +
ALIGN(dest_sz, SZ_64);
ptr = dma_alloc_coherent(__scm->dev, ptr_sz, &ptr_phys, GFP_KERNEL);
if (!ptr)
return -ENOMEM;
/* Fill source vmid detail */
src = ptr;
i = 0;
for (b = 0; b < BITS_PER_TYPE(u64); b++) {
if (srcvm_bits & BIT(b))
src[i++] = cpu_to_le32(b);
}
/* Fill details of mem buff to map */
mem_to_map = ptr + ALIGN(src_sz, SZ_64);
mem_to_map_phys = ptr_phys + ALIGN(src_sz, SZ_64);
mem_to_map->mem_addr = cpu_to_le64(mem_addr);
mem_to_map->mem_size = cpu_to_le64(mem_sz);
next_vm = 0;
/* Fill details of next vmid detail */
destvm = ptr + ALIGN(mem_to_map_sz, SZ_64) + ALIGN(src_sz, SZ_64);
dest_phys = ptr_phys + ALIGN(mem_to_map_sz, SZ_64) + ALIGN(src_sz, SZ_64);
for (i = 0; i < dest_cnt; i++, destvm++, newvm++) {
destvm->vmid = cpu_to_le32(newvm->vmid);
destvm->perm = cpu_to_le32(newvm->perm);
destvm->ctx = 0;
destvm->ctx_size = 0;
next_vm |= BIT(newvm->vmid);
}
ret = __qcom_scm_assign_mem(__scm->dev, mem_to_map_phys, mem_to_map_sz,
ptr_phys, src_sz, dest_phys, dest_sz);
dma_free_coherent(__scm->dev, ptr_sz, ptr, ptr_phys);
if (ret) {
dev_err(__scm->dev,
"Assign memory protection call failed %d\n", ret);
return -EINVAL;
}
*srcvm = next_vm;
return 0;
}
EXPORT_SYMBOL(qcom_scm_assign_mem);
/**
* qcom_scm_ocmem_lock_available() - is OCMEM lock/unlock interface available
*/
bool qcom_scm_ocmem_lock_available(void)
{
return __qcom_scm_is_call_available(__scm->dev, QCOM_SCM_SVC_OCMEM,
QCOM_SCM_OCMEM_LOCK_CMD);
}
EXPORT_SYMBOL(qcom_scm_ocmem_lock_available);
/**
* qcom_scm_ocmem_lock() - call OCMEM lock interface to assign an OCMEM
* region to the specified initiator
*
* @id: tz initiator id
* @offset: OCMEM offset
* @size: OCMEM size
* @mode: access mode (WIDE/NARROW)
*/
int qcom_scm_ocmem_lock(enum qcom_scm_ocmem_client id, u32 offset, u32 size,
u32 mode)
{
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_OCMEM,
.cmd = QCOM_SCM_OCMEM_LOCK_CMD,
.args[0] = id,
.args[1] = offset,
.args[2] = size,
.args[3] = mode,
.arginfo = QCOM_SCM_ARGS(4),
};
return qcom_scm_call(__scm->dev, &desc, NULL);
}
EXPORT_SYMBOL(qcom_scm_ocmem_lock);
/**
* qcom_scm_ocmem_unlock() - call OCMEM unlock interface to release an OCMEM
* region from the specified initiator
*
* @id: tz initiator id
* @offset: OCMEM offset
* @size: OCMEM size
*/
int qcom_scm_ocmem_unlock(enum qcom_scm_ocmem_client id, u32 offset, u32 size)
{
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_OCMEM,
.cmd = QCOM_SCM_OCMEM_UNLOCK_CMD,
.args[0] = id,
.args[1] = offset,
.args[2] = size,
.arginfo = QCOM_SCM_ARGS(3),
};
return qcom_scm_call(__scm->dev, &desc, NULL);
}
EXPORT_SYMBOL(qcom_scm_ocmem_unlock);
/**
* qcom_scm_ice_available() - Is the ICE key programming interface available?
*
* Return: true iff the SCM calls wrapped by qcom_scm_ice_invalidate_key() and
* qcom_scm_ice_set_key() are available.
*/
bool qcom_scm_ice_available(void)
{
return __qcom_scm_is_call_available(__scm->dev, QCOM_SCM_SVC_ES,
QCOM_SCM_ES_INVALIDATE_ICE_KEY) &&
__qcom_scm_is_call_available(__scm->dev, QCOM_SCM_SVC_ES,
QCOM_SCM_ES_CONFIG_SET_ICE_KEY);
}
EXPORT_SYMBOL(qcom_scm_ice_available);
/**
* qcom_scm_ice_invalidate_key() - Invalidate an inline encryption key
* @index: the keyslot to invalidate
*
* The UFSHCI and eMMC standards define a standard way to do this, but it
* doesn't work on these SoCs; only this SCM call does.
*
* It is assumed that the SoC has only one ICE instance being used, as this SCM
* call doesn't specify which ICE instance the keyslot belongs to.
*
* Return: 0 on success; -errno on failure.
*/
int qcom_scm_ice_invalidate_key(u32 index)
{
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_ES,
.cmd = QCOM_SCM_ES_INVALIDATE_ICE_KEY,
.arginfo = QCOM_SCM_ARGS(1),
.args[0] = index,
.owner = ARM_SMCCC_OWNER_SIP,
};
return qcom_scm_call(__scm->dev, &desc, NULL);
}
EXPORT_SYMBOL(qcom_scm_ice_invalidate_key);
/**
* qcom_scm_ice_set_key() - Set an inline encryption key
* @index: the keyslot into which to set the key
* @key: the key to program
* @key_size: the size of the key in bytes
* @cipher: the encryption algorithm the key is for
* @data_unit_size: the encryption data unit size, i.e. the size of each
* individual plaintext and ciphertext. Given in 512-byte
* units, e.g. 1 = 512 bytes, 8 = 4096 bytes, etc.
*
* Program a key into a keyslot of Qualcomm ICE (Inline Crypto Engine), where it
* can then be used to encrypt/decrypt UFS or eMMC I/O requests inline.
*
* The UFSHCI and eMMC standards define a standard way to do this, but it
* doesn't work on these SoCs; only this SCM call does.
*
* It is assumed that the SoC has only one ICE instance being used, as this SCM
* call doesn't specify which ICE instance the keyslot belongs to.
*
* Return: 0 on success; -errno on failure.
*/
int qcom_scm_ice_set_key(u32 index, const u8 *key, u32 key_size,
enum qcom_scm_ice_cipher cipher, u32 data_unit_size)
{
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_ES,
.cmd = QCOM_SCM_ES_CONFIG_SET_ICE_KEY,
.arginfo = QCOM_SCM_ARGS(5, QCOM_SCM_VAL, QCOM_SCM_RW,
QCOM_SCM_VAL, QCOM_SCM_VAL,
QCOM_SCM_VAL),
.args[0] = index,
.args[2] = key_size,
.args[3] = cipher,
.args[4] = data_unit_size,
.owner = ARM_SMCCC_OWNER_SIP,
};
void *keybuf;
dma_addr_t key_phys;
int ret;
/*
* 'key' may point to vmalloc()'ed memory, but we need to pass a
* physical address that's been properly flushed. The sanctioned way to
* do this is by using the DMA API. But as is best practice for crypto
* keys, we also must wipe the key after use. This makes kmemdup() +
* dma_map_single() not clearly correct, since the DMA API can use
* bounce buffers. Instead, just use dma_alloc_coherent(). Programming
* keys is normally rare and thus not performance-critical.
*/
keybuf = dma_alloc_coherent(__scm->dev, key_size, &key_phys,
GFP_KERNEL);
if (!keybuf)
return -ENOMEM;
memcpy(keybuf, key, key_size);
desc.args[1] = key_phys;
ret = qcom_scm_call(__scm->dev, &desc, NULL);
memzero_explicit(keybuf, key_size);
dma_free_coherent(__scm->dev, key_size, keybuf, key_phys);
return ret;
}
EXPORT_SYMBOL(qcom_scm_ice_set_key);
/**
* qcom_scm_hdcp_available() - Check if secure environment supports HDCP.
*
* Return true if HDCP is supported, false if not.
*/
bool qcom_scm_hdcp_available(void)
{
bool avail;
int ret = qcom_scm_clk_enable();
if (ret)
return ret;
avail = __qcom_scm_is_call_available(__scm->dev, QCOM_SCM_SVC_HDCP,
QCOM_SCM_HDCP_INVOKE);
qcom_scm_clk_disable();
return avail;
}
EXPORT_SYMBOL(qcom_scm_hdcp_available);
/**
* qcom_scm_hdcp_req() - Send HDCP request.
* @req: HDCP request array
* @req_cnt: HDCP request array count
* @resp: response buffer passed to SCM
*
* Write HDCP register(s) through SCM.
*/
int qcom_scm_hdcp_req(struct qcom_scm_hdcp_req *req, u32 req_cnt, u32 *resp)
{
int ret;
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_HDCP,
.cmd = QCOM_SCM_HDCP_INVOKE,
.arginfo = QCOM_SCM_ARGS(10),
.args = {
req[0].addr,
req[0].val,
req[1].addr,
req[1].val,
req[2].addr,
req[2].val,
req[3].addr,
req[3].val,
req[4].addr,
req[4].val
},
.owner = ARM_SMCCC_OWNER_SIP,
};
struct qcom_scm_res res;
if (req_cnt > QCOM_SCM_HDCP_MAX_REQ_CNT)
return -ERANGE;
ret = qcom_scm_clk_enable();
if (ret)
return ret;
ret = qcom_scm_call(__scm->dev, &desc, &res);
*resp = res.result[0];
qcom_scm_clk_disable();
return ret;
}
EXPORT_SYMBOL(qcom_scm_hdcp_req);
int qcom_scm_iommu_set_pt_format(u32 sec_id, u32 ctx_num, u32 pt_fmt)
{
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_SMMU_PROGRAM,
.cmd = QCOM_SCM_SMMU_PT_FORMAT,
.arginfo = QCOM_SCM_ARGS(3),
.args[0] = sec_id,
.args[1] = ctx_num,
.args[2] = pt_fmt, /* 0: LPAE AArch32 - 1: AArch64 */
.owner = ARM_SMCCC_OWNER_SIP,
};
return qcom_scm_call(__scm->dev, &desc, NULL);
}
EXPORT_SYMBOL(qcom_scm_iommu_set_pt_format);
int qcom_scm_qsmmu500_wait_safe_toggle(bool en)
{
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_SMMU_PROGRAM,
.cmd = QCOM_SCM_SMMU_CONFIG_ERRATA1,
.arginfo = QCOM_SCM_ARGS(2),
.args[0] = QCOM_SCM_SMMU_CONFIG_ERRATA1_CLIENT_ALL,
.args[1] = en,
.owner = ARM_SMCCC_OWNER_SIP,
};
return qcom_scm_call_atomic(__scm->dev, &desc, NULL);
}
EXPORT_SYMBOL(qcom_scm_qsmmu500_wait_safe_toggle);
bool qcom_scm_lmh_dcvsh_available(void)
{
return __qcom_scm_is_call_available(__scm->dev, QCOM_SCM_SVC_LMH, QCOM_SCM_LMH_LIMIT_DCVSH);
}
EXPORT_SYMBOL(qcom_scm_lmh_dcvsh_available);
int qcom_scm_lmh_profile_change(u32 profile_id)
{
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_LMH,
.cmd = QCOM_SCM_LMH_LIMIT_PROFILE_CHANGE,
.arginfo = QCOM_SCM_ARGS(1, QCOM_SCM_VAL),
.args[0] = profile_id,
.owner = ARM_SMCCC_OWNER_SIP,
};
return qcom_scm_call(__scm->dev, &desc, NULL);
}
EXPORT_SYMBOL(qcom_scm_lmh_profile_change);
int qcom_scm_lmh_dcvsh(u32 payload_fn, u32 payload_reg, u32 payload_val,
u64 limit_node, u32 node_id, u64 version)
{
dma_addr_t payload_phys;
u32 *payload_buf;
int ret, payload_size = 5 * sizeof(u32);
struct qcom_scm_desc desc = {
.svc = QCOM_SCM_SVC_LMH,
.cmd = QCOM_SCM_LMH_LIMIT_DCVSH,
.arginfo = QCOM_SCM_ARGS(5, QCOM_SCM_RO, QCOM_SCM_VAL, QCOM_SCM_VAL,
QCOM_SCM_VAL, QCOM_SCM_VAL),
.args[1] = payload_size,
.args[2] = limit_node,
.args[3] = node_id,
.args[4] = version,
.owner = ARM_SMCCC_OWNER_SIP,
};
payload_buf = dma_alloc_coherent(__scm->dev, payload_size, &payload_phys, GFP_KERNEL);
if (!payload_buf)
return -ENOMEM;
payload_buf[0] = payload_fn;
payload_buf[1] = 0;
payload_buf[2] = payload_reg;
payload_buf[3] = 1;
payload_buf[4] = payload_val;
desc.args[0] = payload_phys;
ret = qcom_scm_call(__scm->dev, &desc, NULL);
dma_free_coherent(__scm->dev, payload_size, payload_buf, payload_phys);
return ret;
}
EXPORT_SYMBOL(qcom_scm_lmh_dcvsh);
static int qcom_scm_find_dload_address(struct device *dev, u64 *addr)
{
struct device_node *tcsr;
struct device_node *np = dev->of_node;
struct resource res;
u32 offset;
int ret;
tcsr = of_parse_phandle(np, "qcom,dload-mode", 0);
if (!tcsr)
return 0;
ret = of_address_to_resource(tcsr, 0, &res);
of_node_put(tcsr);
if (ret)
return ret;
ret = of_property_read_u32_index(np, "qcom,dload-mode", 1, &offset);
if (ret < 0)
return ret;
*addr = res.start + offset;
return 0;
}
/**
* qcom_scm_is_available() - Checks if SCM is available
*/
bool qcom_scm_is_available(void)
{
return !!__scm;
}
EXPORT_SYMBOL(qcom_scm_is_available);
firmware: qcom: scm: Add wait-queue handling logic When the firmware (FW) supports multiple requests per VM, multiple requests from the same/different VM can reach the firmware at the same time. Since the firmware currently being used has limited resources, it guards them with a resource lock and puts requests on a wait-queue internally and signals to HLOS that it is doing so. It does this by returning a new return value in addition to success or error: SCM_WAITQ_SLEEP. A sleeping SCM call can be woken up by an interrupt that the FW raises. 1) SCM_WAITQ_SLEEP: When an SCM call receives this return value instead of success or error, FW has placed this call on a wait-queue and has signalled HLOS to put it to non-interruptible sleep. Along with this return value, FW also passes to HLOS `wq_ctx` - a unique number (UID) identifying the wait-queue that it has put the call on, internally. This is to help HLOS with its own bookkeeping to wake this sleeping call later. Additionally, FW also passes to HLOS `smc_call_ctx` - a UID identifying the SCM call thus being put to sleep. This is also for HLOS' bookkeeping to wake this call up later. These two additional values are passed via the a1 and a2 registers. N.B.: The "ctx" in the above UID names = "context". The handshake mechanism that HLOS uses to talk to FW about wait-queue operations involves two new SMC calls. 1) get_wq_ctx(): Arguments: None Returns: wq_ctx, flags, more_pending Get the wait-queue context, and wake up either one or all of the sleeping SCM calls associated with that wait-queue. Additionally, repeat this if there are more wait-queues that are ready to have their requests woken up (`more_pending`). 2) wq_resume(smc_call_ctx): Arguments: smc_call_ctx HLOS needs to issue this in response to receiving an IRQ, passing to FW the same smc_call_ctx that FW receives from HLOS via the get_wq_ctx() call. (The mechanism to wake a SMC call back up is described in detail below) VM_1 VM_2 Firmware │ │ │ │ │ │ │ │ │ │ │ │ │ REQUEST_1 │ │ ├────────────────────────┼─────────────────────────────────┤ │ │ │ │ │ ┌──┼──┐ │ │ │ │ │ │ │ REQUEST_2 │ │ │ │ ├──────────────────────────────┼──┤ │ │ │ │ │ │Resource │ │ │ │ │is busy │ │ {WQ_SLEEP} │ │ │ │ │◄─────────────────────────────┼──┤ │ │ │ wq_ctx, smc_call_ctx │ │ │ │ │ └──┼──┘ │ REQUEST_1 COMPLETE │ │ │◄───────────────────────┼─────────────────────────────────┤ │ │ │ │ │ IRQ │ │ │◄─-------------------------------│ │ │ │ │ │ get_wq_ctx() │ │ ├────────────────────────────────►│ │ │ │ │ │ │ │ │◄────────────────────────────────┤ │ │ wq_ctx, flags, and │ │ │ more_pending │ │ │ │ │ │ │ │ │ wq_resume(smc_call_ctx) │ │ ├────────────────────────────────►│ │ │ │ │ │ │ │ │ REQUEST_2 COMPLETE │ │ │◄────────────────────────────────┤ │ │ │ │ │ │ With the exception of get_wq_ctx(), the other SMC call wq_resume() can return WQ_SLEEP (these nested rounds of WQ_SLEEP are not shown in the above diagram for the sake of simplicity). Therefore, introduce a new do-while loop to handle multiple WQ_SLEEP return values for the same parent SCM call. Request Completion in the above diagram refers to either a success return value (zero) or error (and not SMC_WAITQ_SLEEP) Also add the interrupt handler that wakes up a sleeping SCM call. Signed-off-by: Guru Das Srinagesh <quic_gurus@quicinc.com> Co-developed-by: Sibi Sankar <quic_sibis@quicinc.com> Signed-off-by: Sibi Sankar <quic_sibis@quicinc.com> Reviewed-by: Guru Das Srinagesh <quic_gurus@quicinc.com> Signed-off-by: Bjorn Andersson <andersson@kernel.org> Link: https://lore.kernel.org/r/20230113161114.22607-3-quic_sibis@quicinc.com
2023-01-14 00:11:14 +08:00
static int qcom_scm_assert_valid_wq_ctx(u32 wq_ctx)
{
/* FW currently only supports a single wq_ctx (zero).
* TODO: Update this logic to include dynamic allocation and lookup of
* completion structs when FW supports more wq_ctx values.
*/
if (wq_ctx != 0) {
dev_err(__scm->dev, "Firmware unexpectedly passed non-zero wq_ctx\n");
return -EINVAL;
}
return 0;
}
int qcom_scm_wait_for_wq_completion(u32 wq_ctx)
{
int ret;
ret = qcom_scm_assert_valid_wq_ctx(wq_ctx);
if (ret)
return ret;
wait_for_completion(&__scm->waitq_comp);
return 0;
}
static int qcom_scm_waitq_wakeup(struct qcom_scm *scm, unsigned int wq_ctx)
{
int ret;
ret = qcom_scm_assert_valid_wq_ctx(wq_ctx);
if (ret)
return ret;
complete(&__scm->waitq_comp);
return 0;
}
static irqreturn_t qcom_scm_irq_handler(int irq, void *data)
{
int ret;
struct qcom_scm *scm = data;
u32 wq_ctx, flags, more_pending = 0;
do {
ret = scm_get_wq_ctx(&wq_ctx, &flags, &more_pending);
if (ret) {
dev_err(scm->dev, "GET_WQ_CTX SMC call failed: %d\n", ret);
goto out;
}
if (flags != QCOM_SMC_WAITQ_FLAG_WAKE_ONE &&
flags != QCOM_SMC_WAITQ_FLAG_WAKE_ALL) {
dev_err(scm->dev, "Invalid flags found for wq_ctx: %u\n", flags);
goto out;
}
ret = qcom_scm_waitq_wakeup(scm, wq_ctx);
if (ret)
goto out;
} while (more_pending);
out:
return IRQ_HANDLED;
}
static int qcom_scm_probe(struct platform_device *pdev)
{
struct qcom_scm *scm;
unsigned long clks;
firmware: qcom: scm: Add wait-queue handling logic When the firmware (FW) supports multiple requests per VM, multiple requests from the same/different VM can reach the firmware at the same time. Since the firmware currently being used has limited resources, it guards them with a resource lock and puts requests on a wait-queue internally and signals to HLOS that it is doing so. It does this by returning a new return value in addition to success or error: SCM_WAITQ_SLEEP. A sleeping SCM call can be woken up by an interrupt that the FW raises. 1) SCM_WAITQ_SLEEP: When an SCM call receives this return value instead of success or error, FW has placed this call on a wait-queue and has signalled HLOS to put it to non-interruptible sleep. Along with this return value, FW also passes to HLOS `wq_ctx` - a unique number (UID) identifying the wait-queue that it has put the call on, internally. This is to help HLOS with its own bookkeeping to wake this sleeping call later. Additionally, FW also passes to HLOS `smc_call_ctx` - a UID identifying the SCM call thus being put to sleep. This is also for HLOS' bookkeeping to wake this call up later. These two additional values are passed via the a1 and a2 registers. N.B.: The "ctx" in the above UID names = "context". The handshake mechanism that HLOS uses to talk to FW about wait-queue operations involves two new SMC calls. 1) get_wq_ctx(): Arguments: None Returns: wq_ctx, flags, more_pending Get the wait-queue context, and wake up either one or all of the sleeping SCM calls associated with that wait-queue. Additionally, repeat this if there are more wait-queues that are ready to have their requests woken up (`more_pending`). 2) wq_resume(smc_call_ctx): Arguments: smc_call_ctx HLOS needs to issue this in response to receiving an IRQ, passing to FW the same smc_call_ctx that FW receives from HLOS via the get_wq_ctx() call. (The mechanism to wake a SMC call back up is described in detail below) VM_1 VM_2 Firmware │ │ │ │ │ │ │ │ │ │ │ │ │ REQUEST_1 │ │ ├────────────────────────┼─────────────────────────────────┤ │ │ │ │ │ ┌──┼──┐ │ │ │ │ │ │ │ REQUEST_2 │ │ │ │ ├──────────────────────────────┼──┤ │ │ │ │ │ │Resource │ │ │ │ │is busy │ │ {WQ_SLEEP} │ │ │ │ │◄─────────────────────────────┼──┤ │ │ │ wq_ctx, smc_call_ctx │ │ │ │ │ └──┼──┘ │ REQUEST_1 COMPLETE │ │ │◄───────────────────────┼─────────────────────────────────┤ │ │ │ │ │ IRQ │ │ │◄─-------------------------------│ │ │ │ │ │ get_wq_ctx() │ │ ├────────────────────────────────►│ │ │ │ │ │ │ │ │◄────────────────────────────────┤ │ │ wq_ctx, flags, and │ │ │ more_pending │ │ │ │ │ │ │ │ │ wq_resume(smc_call_ctx) │ │ ├────────────────────────────────►│ │ │ │ │ │ │ │ │ REQUEST_2 COMPLETE │ │ │◄────────────────────────────────┤ │ │ │ │ │ │ With the exception of get_wq_ctx(), the other SMC call wq_resume() can return WQ_SLEEP (these nested rounds of WQ_SLEEP are not shown in the above diagram for the sake of simplicity). Therefore, introduce a new do-while loop to handle multiple WQ_SLEEP return values for the same parent SCM call. Request Completion in the above diagram refers to either a success return value (zero) or error (and not SMC_WAITQ_SLEEP) Also add the interrupt handler that wakes up a sleeping SCM call. Signed-off-by: Guru Das Srinagesh <quic_gurus@quicinc.com> Co-developed-by: Sibi Sankar <quic_sibis@quicinc.com> Signed-off-by: Sibi Sankar <quic_sibis@quicinc.com> Reviewed-by: Guru Das Srinagesh <quic_gurus@quicinc.com> Signed-off-by: Bjorn Andersson <andersson@kernel.org> Link: https://lore.kernel.org/r/20230113161114.22607-3-quic_sibis@quicinc.com
2023-01-14 00:11:14 +08:00
int irq, ret;
scm = devm_kzalloc(&pdev->dev, sizeof(*scm), GFP_KERNEL);
if (!scm)
return -ENOMEM;
ret = qcom_scm_find_dload_address(&pdev->dev, &scm->dload_mode_addr);
if (ret < 0)
return ret;
mutex_init(&scm->scm_bw_lock);
clks = (unsigned long)of_device_get_match_data(&pdev->dev);
scm->path = devm_of_icc_get(&pdev->dev, NULL);
if (IS_ERR(scm->path))
return dev_err_probe(&pdev->dev, PTR_ERR(scm->path),
"failed to acquire interconnect path\n");
scm->core_clk = devm_clk_get(&pdev->dev, "core");
if (IS_ERR(scm->core_clk)) {
if (PTR_ERR(scm->core_clk) == -EPROBE_DEFER)
return PTR_ERR(scm->core_clk);
if (clks & SCM_HAS_CORE_CLK) {
dev_err(&pdev->dev, "failed to acquire core clk\n");
return PTR_ERR(scm->core_clk);
}
scm->core_clk = NULL;
}
scm->iface_clk = devm_clk_get(&pdev->dev, "iface");
if (IS_ERR(scm->iface_clk)) {
if (PTR_ERR(scm->iface_clk) == -EPROBE_DEFER)
return PTR_ERR(scm->iface_clk);
if (clks & SCM_HAS_IFACE_CLK) {
dev_err(&pdev->dev, "failed to acquire iface clk\n");
return PTR_ERR(scm->iface_clk);
}
scm->iface_clk = NULL;
}
scm->bus_clk = devm_clk_get(&pdev->dev, "bus");
if (IS_ERR(scm->bus_clk)) {
if (PTR_ERR(scm->bus_clk) == -EPROBE_DEFER)
return PTR_ERR(scm->bus_clk);
if (clks & SCM_HAS_BUS_CLK) {
dev_err(&pdev->dev, "failed to acquire bus clk\n");
return PTR_ERR(scm->bus_clk);
}
scm->bus_clk = NULL;
}
scm->reset.ops = &qcom_scm_pas_reset_ops;
scm->reset.nr_resets = 1;
scm->reset.of_node = pdev->dev.of_node;
ret = devm_reset_controller_register(&pdev->dev, &scm->reset);
if (ret)
return ret;
/* vote for max clk rate for highest performance */
ret = clk_set_rate(scm->core_clk, INT_MAX);
if (ret)
return ret;
__scm = scm;
__scm->dev = &pdev->dev;
firmware: qcom: scm: Add wait-queue handling logic When the firmware (FW) supports multiple requests per VM, multiple requests from the same/different VM can reach the firmware at the same time. Since the firmware currently being used has limited resources, it guards them with a resource lock and puts requests on a wait-queue internally and signals to HLOS that it is doing so. It does this by returning a new return value in addition to success or error: SCM_WAITQ_SLEEP. A sleeping SCM call can be woken up by an interrupt that the FW raises. 1) SCM_WAITQ_SLEEP: When an SCM call receives this return value instead of success or error, FW has placed this call on a wait-queue and has signalled HLOS to put it to non-interruptible sleep. Along with this return value, FW also passes to HLOS `wq_ctx` - a unique number (UID) identifying the wait-queue that it has put the call on, internally. This is to help HLOS with its own bookkeeping to wake this sleeping call later. Additionally, FW also passes to HLOS `smc_call_ctx` - a UID identifying the SCM call thus being put to sleep. This is also for HLOS' bookkeeping to wake this call up later. These two additional values are passed via the a1 and a2 registers. N.B.: The "ctx" in the above UID names = "context". The handshake mechanism that HLOS uses to talk to FW about wait-queue operations involves two new SMC calls. 1) get_wq_ctx(): Arguments: None Returns: wq_ctx, flags, more_pending Get the wait-queue context, and wake up either one or all of the sleeping SCM calls associated with that wait-queue. Additionally, repeat this if there are more wait-queues that are ready to have their requests woken up (`more_pending`). 2) wq_resume(smc_call_ctx): Arguments: smc_call_ctx HLOS needs to issue this in response to receiving an IRQ, passing to FW the same smc_call_ctx that FW receives from HLOS via the get_wq_ctx() call. (The mechanism to wake a SMC call back up is described in detail below) VM_1 VM_2 Firmware │ │ │ │ │ │ │ │ │ │ │ │ │ REQUEST_1 │ │ ├────────────────────────┼─────────────────────────────────┤ │ │ │ │ │ ┌──┼──┐ │ │ │ │ │ │ │ REQUEST_2 │ │ │ │ ├──────────────────────────────┼──┤ │ │ │ │ │ │Resource │ │ │ │ │is busy │ │ {WQ_SLEEP} │ │ │ │ │◄─────────────────────────────┼──┤ │ │ │ wq_ctx, smc_call_ctx │ │ │ │ │ └──┼──┘ │ REQUEST_1 COMPLETE │ │ │◄───────────────────────┼─────────────────────────────────┤ │ │ │ │ │ IRQ │ │ │◄─-------------------------------│ │ │ │ │ │ get_wq_ctx() │ │ ├────────────────────────────────►│ │ │ │ │ │ │ │ │◄────────────────────────────────┤ │ │ wq_ctx, flags, and │ │ │ more_pending │ │ │ │ │ │ │ │ │ wq_resume(smc_call_ctx) │ │ ├────────────────────────────────►│ │ │ │ │ │ │ │ │ REQUEST_2 COMPLETE │ │ │◄────────────────────────────────┤ │ │ │ │ │ │ With the exception of get_wq_ctx(), the other SMC call wq_resume() can return WQ_SLEEP (these nested rounds of WQ_SLEEP are not shown in the above diagram for the sake of simplicity). Therefore, introduce a new do-while loop to handle multiple WQ_SLEEP return values for the same parent SCM call. Request Completion in the above diagram refers to either a success return value (zero) or error (and not SMC_WAITQ_SLEEP) Also add the interrupt handler that wakes up a sleeping SCM call. Signed-off-by: Guru Das Srinagesh <quic_gurus@quicinc.com> Co-developed-by: Sibi Sankar <quic_sibis@quicinc.com> Signed-off-by: Sibi Sankar <quic_sibis@quicinc.com> Reviewed-by: Guru Das Srinagesh <quic_gurus@quicinc.com> Signed-off-by: Bjorn Andersson <andersson@kernel.org> Link: https://lore.kernel.org/r/20230113161114.22607-3-quic_sibis@quicinc.com
2023-01-14 00:11:14 +08:00
init_completion(&__scm->waitq_comp);
irq = platform_get_irq_optional(pdev, 0);
firmware: qcom: scm: Add wait-queue handling logic When the firmware (FW) supports multiple requests per VM, multiple requests from the same/different VM can reach the firmware at the same time. Since the firmware currently being used has limited resources, it guards them with a resource lock and puts requests on a wait-queue internally and signals to HLOS that it is doing so. It does this by returning a new return value in addition to success or error: SCM_WAITQ_SLEEP. A sleeping SCM call can be woken up by an interrupt that the FW raises. 1) SCM_WAITQ_SLEEP: When an SCM call receives this return value instead of success or error, FW has placed this call on a wait-queue and has signalled HLOS to put it to non-interruptible sleep. Along with this return value, FW also passes to HLOS `wq_ctx` - a unique number (UID) identifying the wait-queue that it has put the call on, internally. This is to help HLOS with its own bookkeeping to wake this sleeping call later. Additionally, FW also passes to HLOS `smc_call_ctx` - a UID identifying the SCM call thus being put to sleep. This is also for HLOS' bookkeeping to wake this call up later. These two additional values are passed via the a1 and a2 registers. N.B.: The "ctx" in the above UID names = "context". The handshake mechanism that HLOS uses to talk to FW about wait-queue operations involves two new SMC calls. 1) get_wq_ctx(): Arguments: None Returns: wq_ctx, flags, more_pending Get the wait-queue context, and wake up either one or all of the sleeping SCM calls associated with that wait-queue. Additionally, repeat this if there are more wait-queues that are ready to have their requests woken up (`more_pending`). 2) wq_resume(smc_call_ctx): Arguments: smc_call_ctx HLOS needs to issue this in response to receiving an IRQ, passing to FW the same smc_call_ctx that FW receives from HLOS via the get_wq_ctx() call. (The mechanism to wake a SMC call back up is described in detail below) VM_1 VM_2 Firmware │ │ │ │ │ │ │ │ │ │ │ │ │ REQUEST_1 │ │ ├────────────────────────┼─────────────────────────────────┤ │ │ │ │ │ ┌──┼──┐ │ │ │ │ │ │ │ REQUEST_2 │ │ │ │ ├──────────────────────────────┼──┤ │ │ │ │ │ │Resource │ │ │ │ │is busy │ │ {WQ_SLEEP} │ │ │ │ │◄─────────────────────────────┼──┤ │ │ │ wq_ctx, smc_call_ctx │ │ │ │ │ └──┼──┘ │ REQUEST_1 COMPLETE │ │ │◄───────────────────────┼─────────────────────────────────┤ │ │ │ │ │ IRQ │ │ │◄─-------------------------------│ │ │ │ │ │ get_wq_ctx() │ │ ├────────────────────────────────►│ │ │ │ │ │ │ │ │◄────────────────────────────────┤ │ │ wq_ctx, flags, and │ │ │ more_pending │ │ │ │ │ │ │ │ │ wq_resume(smc_call_ctx) │ │ ├────────────────────────────────►│ │ │ │ │ │ │ │ │ REQUEST_2 COMPLETE │ │ │◄────────────────────────────────┤ │ │ │ │ │ │ With the exception of get_wq_ctx(), the other SMC call wq_resume() can return WQ_SLEEP (these nested rounds of WQ_SLEEP are not shown in the above diagram for the sake of simplicity). Therefore, introduce a new do-while loop to handle multiple WQ_SLEEP return values for the same parent SCM call. Request Completion in the above diagram refers to either a success return value (zero) or error (and not SMC_WAITQ_SLEEP) Also add the interrupt handler that wakes up a sleeping SCM call. Signed-off-by: Guru Das Srinagesh <quic_gurus@quicinc.com> Co-developed-by: Sibi Sankar <quic_sibis@quicinc.com> Signed-off-by: Sibi Sankar <quic_sibis@quicinc.com> Reviewed-by: Guru Das Srinagesh <quic_gurus@quicinc.com> Signed-off-by: Bjorn Andersson <andersson@kernel.org> Link: https://lore.kernel.org/r/20230113161114.22607-3-quic_sibis@quicinc.com
2023-01-14 00:11:14 +08:00
if (irq < 0) {
if (irq != -ENXIO)
return irq;
} else {
ret = devm_request_threaded_irq(__scm->dev, irq, NULL, qcom_scm_irq_handler,
IRQF_ONESHOT, "qcom-scm", __scm);
if (ret < 0)
return dev_err_probe(scm->dev, ret, "Failed to request qcom-scm irq\n");
}
firmware: qcom_scm: Reduce locking section for __get_convention() We shouldn't need to hold this spinlock here around the entire SCM call into the firmware and back. Instead, we should be able to query the firmware, potentially in parallel with other CPUs making the same convention detection firmware call, and then grab the lock to update the calling convention detected. The convention doesn't change at runtime so calling into firmware more than once is possibly wasteful but simpler. Besides, this is the slow path, not the fast path where we've already detected the convention used. More importantly, this allows us to add more logic here to workaround the case where the firmware call to check for availability isn't implemented in the firmware at all. In that case we can check the firmware node compatible string and force a calling convention. Note that we remove the 'has_queried' logic that is repeated twice. That could lead to the calling convention being printed multiple times to the kernel logs if the bool is true but __query_convention() is running on multiple CPUs. We also shorten the time where the lock is held, but we keep the lock held around the printk because it doesn't seem hugely important to drop it for that. Cc: Elliot Berman <eberman@codeaurora.org> Cc: Brian Masney <masneyb@onstation.org> Cc: Stephan Gerhold <stephan@gerhold.net> Cc: Jeffrey Hugo <jhugo@codeaurora.org> Cc: Douglas Anderson <dianders@chromium.org> Fixes: 9a434cee773a ("firmware: qcom_scm: Dynamically support SMCCC and legacy conventions") Signed-off-by: Stephen Boyd <swboyd@chromium.org> Link: https://lore.kernel.org/r/20210223214539.1336155-3-swboyd@chromium.org Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
2021-02-24 05:45:35 +08:00
__get_convention();
/*
* If requested enable "download mode", from this point on warmboot
* will cause the boot stages to enter download mode, unless
* disabled below by a clean shutdown/reboot.
*/
if (download_mode)
qcom_scm_set_download_mode(true);
return 0;
}
static void qcom_scm_shutdown(struct platform_device *pdev)
{
/* Clean shutdown, disable download mode to allow normal restart */
qcom_scm_set_download_mode(false);
}
static const struct of_device_id qcom_scm_dt_match[] = {
{ .compatible = "qcom,scm-apq8064",
/* FIXME: This should have .data = (void *) SCM_HAS_CORE_CLK */
},
{ .compatible = "qcom,scm-apq8084", .data = (void *)(SCM_HAS_CORE_CLK |
SCM_HAS_IFACE_CLK |
SCM_HAS_BUS_CLK)
},
{ .compatible = "qcom,scm-ipq4019" },
{ .compatible = "qcom,scm-mdm9607", .data = (void *)(SCM_HAS_CORE_CLK |
SCM_HAS_IFACE_CLK |
SCM_HAS_BUS_CLK) },
{ .compatible = "qcom,scm-msm8660", .data = (void *) SCM_HAS_CORE_CLK },
{ .compatible = "qcom,scm-msm8960", .data = (void *) SCM_HAS_CORE_CLK },
{ .compatible = "qcom,scm-msm8916", .data = (void *)(SCM_HAS_CORE_CLK |
SCM_HAS_IFACE_CLK |
SCM_HAS_BUS_CLK)
},
{ .compatible = "qcom,scm-msm8953", .data = (void *)(SCM_HAS_CORE_CLK |
SCM_HAS_IFACE_CLK |
SCM_HAS_BUS_CLK)
},
{ .compatible = "qcom,scm-msm8974", .data = (void *)(SCM_HAS_CORE_CLK |
SCM_HAS_IFACE_CLK |
SCM_HAS_BUS_CLK)
},
{ .compatible = "qcom,scm-msm8976", .data = (void *)(SCM_HAS_CORE_CLK |
SCM_HAS_IFACE_CLK |
SCM_HAS_BUS_CLK)
},
{ .compatible = "qcom,scm-msm8994" },
{ .compatible = "qcom,scm-msm8996" },
{ .compatible = "qcom,scm-sm6375", .data = (void *)SCM_HAS_CORE_CLK },
{ .compatible = "qcom,scm" },
{}
};
MODULE_DEVICE_TABLE(of, qcom_scm_dt_match);
static struct platform_driver qcom_scm_driver = {
.driver = {
.name = "qcom_scm",
.of_match_table = qcom_scm_dt_match,
.suppress_bind_attrs = true,
},
.probe = qcom_scm_probe,
.shutdown = qcom_scm_shutdown,
};
static int __init qcom_scm_init(void)
{
return platform_driver_register(&qcom_scm_driver);
}
subsys_initcall(qcom_scm_init);
MODULE_DESCRIPTION("Qualcomm Technologies, Inc. SCM driver");
MODULE_LICENSE("GPL v2");