* [PATCH V2 1/6] cxl: Remove unused values in bare-metal environment.
2017-03-14 11:08 [PATCH V2 0/6] cxl: Add support for Coherent Accelerator Interface Architecture 2.0 Christophe Lombard
@ 2017-03-14 11:08 ` Christophe Lombard
2017-03-20 1:11 ` Andrew Donnellan
` (2 more replies)
2017-03-14 11:08 ` [PATCH V2 2/6] cxl: Keep track of mm struct associated with a context Christophe Lombard
` (5 subsequent siblings)
6 siblings, 3 replies; 20+ messages in thread
From: Christophe Lombard @ 2017-03-14 11:08 UTC (permalink / raw)
To: linuxppc-dev, fbarrat, imunsie, andrew.donnellan
The two fields pid and tid of the structure cxl_irq_info are only used
in the guest environment. To avoid confusion, it's not necessary
to fill the fields in the bare-metal environment.
The PSL Process and Thread Identification Register is only used when
attaching a dedicated process for PSL8 only.
Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com>
---
drivers/misc/cxl/native.c | 5 -----
1 file changed, 5 deletions(-)
diff --git a/drivers/misc/cxl/native.c b/drivers/misc/cxl/native.c
index 7ae7105..7257e8b 100644
--- a/drivers/misc/cxl/native.c
+++ b/drivers/misc/cxl/native.c
@@ -859,8 +859,6 @@ static int native_detach_process(struct cxl_context *ctx)
static int native_get_irq_info(struct cxl_afu *afu, struct cxl_irq_info *info)
{
- u64 pidtid;
-
/* If the adapter has gone away, we can't get any meaningful
* information.
*/
@@ -870,9 +868,6 @@ static int native_get_irq_info(struct cxl_afu *afu, struct cxl_irq_info *info)
info->dsisr = cxl_p2n_read(afu, CXL_PSL_DSISR_An);
info->dar = cxl_p2n_read(afu, CXL_PSL_DAR_An);
info->dsr = cxl_p2n_read(afu, CXL_PSL_DSR_An);
- pidtid = cxl_p2n_read(afu, CXL_PSL_PID_TID_An);
- info->pid = pidtid >> 32;
- info->tid = pidtid & 0xffffffff;
info->afu_err = cxl_p2n_read(afu, CXL_AFU_ERR_An);
info->errstat = cxl_p2n_read(afu, CXL_PSL_ErrStat_An);
info->proc_handle = 0;
--
2.7.4
^ permalink raw reply related [flat|nested] 20+ messages in thread* Re: [PATCH V2 1/6] cxl: Remove unused values in bare-metal environment.
2017-03-14 11:08 ` [PATCH V2 1/6] cxl: Remove unused values in bare-metal environment Christophe Lombard
@ 2017-03-20 1:11 ` Andrew Donnellan
2017-03-20 14:18 ` Frederic Barrat
2017-03-21 11:16 ` Frederic Barrat
2 siblings, 0 replies; 20+ messages in thread
From: Andrew Donnellan @ 2017-03-20 1:11 UTC (permalink / raw)
To: Christophe Lombard, linuxppc-dev, fbarrat, imunsie
Reviewed-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com>
On 14/03/17 22:08, Christophe Lombard wrote:
> The two fields pid and tid of the structure cxl_irq_info are only used
> in the guest environment. To avoid confusion, it's not necessary
> to fill the fields in the bare-metal environment.
> The PSL Process and Thread Identification Register is only used when
> attaching a dedicated process for PSL8 only.
>
> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com>
> ---
> drivers/misc/cxl/native.c | 5 -----
> 1 file changed, 5 deletions(-)
>
> diff --git a/drivers/misc/cxl/native.c b/drivers/misc/cxl/native.c
> index 7ae7105..7257e8b 100644
> --- a/drivers/misc/cxl/native.c
> +++ b/drivers/misc/cxl/native.c
> @@ -859,8 +859,6 @@ static int native_detach_process(struct cxl_context *ctx)
>
> static int native_get_irq_info(struct cxl_afu *afu, struct cxl_irq_info *info)
> {
> - u64 pidtid;
> -
> /* If the adapter has gone away, we can't get any meaningful
> * information.
> */
> @@ -870,9 +868,6 @@ static int native_get_irq_info(struct cxl_afu *afu, struct cxl_irq_info *info)
> info->dsisr = cxl_p2n_read(afu, CXL_PSL_DSISR_An);
> info->dar = cxl_p2n_read(afu, CXL_PSL_DAR_An);
> info->dsr = cxl_p2n_read(afu, CXL_PSL_DSR_An);
> - pidtid = cxl_p2n_read(afu, CXL_PSL_PID_TID_An);
> - info->pid = pidtid >> 32;
> - info->tid = pidtid & 0xffffffff;
> info->afu_err = cxl_p2n_read(afu, CXL_AFU_ERR_An);
> info->errstat = cxl_p2n_read(afu, CXL_PSL_ErrStat_An);
> info->proc_handle = 0;
>
--
Andrew Donnellan OzLabs, ADL Canberra
andrew.donnellan@au1.ibm.com IBM Australia Limited
^ permalink raw reply [flat|nested] 20+ messages in thread* Re: [PATCH V2 1/6] cxl: Remove unused values in bare-metal environment.
2017-03-14 11:08 ` [PATCH V2 1/6] cxl: Remove unused values in bare-metal environment Christophe Lombard
2017-03-20 1:11 ` Andrew Donnellan
@ 2017-03-20 14:18 ` Frederic Barrat
2017-03-21 11:16 ` Frederic Barrat
2 siblings, 0 replies; 20+ messages in thread
From: Frederic Barrat @ 2017-03-20 14:18 UTC (permalink / raw)
To: Christophe Lombard, linuxppc-dev, imunsie, andrew.donnellan
Le 14/03/2017 à 12:08, Christophe Lombard a écrit :
> The two fields pid and tid of the structure cxl_irq_info are only used
> in the guest environment. To avoid confusion, it's not necessary
> to fill the fields in the bare-metal environment.
> The PSL Process and Thread Identification Register is only used when
> attaching a dedicated process for PSL8 only.
>
> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com>
> ---
> drivers/misc/cxl/native.c | 5 -----
> 1 file changed, 5 deletions(-)
>
> diff --git a/drivers/misc/cxl/native.c b/drivers/misc/cxl/native.c
> index 7ae7105..7257e8b 100644
> --- a/drivers/misc/cxl/native.c
> +++ b/drivers/misc/cxl/native.c
> @@ -859,8 +859,6 @@ static int native_detach_process(struct cxl_context *ctx)
>
> static int native_get_irq_info(struct cxl_afu *afu, struct cxl_irq_info *info)
> {
> - u64 pidtid;
> -
> /* If the adapter has gone away, we can't get any meaningful
> * information.
> */
> @@ -870,9 +868,6 @@ static int native_get_irq_info(struct cxl_afu *afu, struct cxl_irq_info *info)
> info->dsisr = cxl_p2n_read(afu, CXL_PSL_DSISR_An);
> info->dar = cxl_p2n_read(afu, CXL_PSL_DAR_An);
> info->dsr = cxl_p2n_read(afu, CXL_PSL_DSR_An);
> - pidtid = cxl_p2n_read(afu, CXL_PSL_PID_TID_An);
> - info->pid = pidtid >> 32;
> - info->tid = pidtid & 0xffffffff;
> info->afu_err = cxl_p2n_read(afu, CXL_AFU_ERR_An);
> info->errstat = cxl_p2n_read(afu, CXL_PSL_ErrStat_An);
> info->proc_handle = 0;
>
Can we rename pid and tid to "reserved" in the struct cxl_irq_info?
By keeping them, they may be used in common code (bare-metal and guest),
which will lead to undefined behavior on bare-metal.
Since the structure matches what is passed in a hcall, we can't get rid
of them, but renaming would be safer.
Fred
^ permalink raw reply [flat|nested] 20+ messages in thread* Re: [PATCH V2 1/6] cxl: Remove unused values in bare-metal environment.
2017-03-14 11:08 ` [PATCH V2 1/6] cxl: Remove unused values in bare-metal environment Christophe Lombard
2017-03-20 1:11 ` Andrew Donnellan
2017-03-20 14:18 ` Frederic Barrat
@ 2017-03-21 11:16 ` Frederic Barrat
2 siblings, 0 replies; 20+ messages in thread
From: Frederic Barrat @ 2017-03-21 11:16 UTC (permalink / raw)
To: linuxppc-dev
Le 14/03/2017 à 12:08, Christophe Lombard a écrit :
> The two fields pid and tid of the structure cxl_irq_info are only used
> in the guest environment. To avoid confusion, it's not necessary
> to fill the fields in the bare-metal environment.
> The PSL Process and Thread Identification Register is only used when
> attaching a dedicated process for PSL8 only.
I forgot to mention: we should probably say in the commit message that
the code change is justified because the CXL_PSL_PID_TID_An register
goes away in CAIA2 and cxl was not really using it.
Fred
>
> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com>
> ---
> drivers/misc/cxl/native.c | 5 -----
> 1 file changed, 5 deletions(-)
>
> diff --git a/drivers/misc/cxl/native.c b/drivers/misc/cxl/native.c
> index 7ae7105..7257e8b 100644
> --- a/drivers/misc/cxl/native.c
> +++ b/drivers/misc/cxl/native.c
> @@ -859,8 +859,6 @@ static int native_detach_process(struct cxl_context *ctx)
>
> static int native_get_irq_info(struct cxl_afu *afu, struct cxl_irq_info *info)
> {
> - u64 pidtid;
> -
> /* If the adapter has gone away, we can't get any meaningful
> * information.
> */
> @@ -870,9 +868,6 @@ static int native_get_irq_info(struct cxl_afu *afu, struct cxl_irq_info *info)
> info->dsisr = cxl_p2n_read(afu, CXL_PSL_DSISR_An);
> info->dar = cxl_p2n_read(afu, CXL_PSL_DAR_An);
> info->dsr = cxl_p2n_read(afu, CXL_PSL_DSR_An);
> - pidtid = cxl_p2n_read(afu, CXL_PSL_PID_TID_An);
> - info->pid = pidtid >> 32;
> - info->tid = pidtid & 0xffffffff;
> info->afu_err = cxl_p2n_read(afu, CXL_AFU_ERR_An);
> info->errstat = cxl_p2n_read(afu, CXL_PSL_ErrStat_An);
> info->proc_handle = 0;
>
^ permalink raw reply [flat|nested] 20+ messages in thread
* [PATCH V2 2/6] cxl: Keep track of mm struct associated with a context
2017-03-14 11:08 [PATCH V2 0/6] cxl: Add support for Coherent Accelerator Interface Architecture 2.0 Christophe Lombard
2017-03-14 11:08 ` [PATCH V2 1/6] cxl: Remove unused values in bare-metal environment Christophe Lombard
@ 2017-03-14 11:08 ` Christophe Lombard
2017-03-20 14:27 ` Frederic Barrat
2017-03-21 11:28 ` Frederic Barrat
2017-03-14 11:08 ` [PATCH V2 3/6] cxl: Update implementation service layer Christophe Lombard
` (4 subsequent siblings)
6 siblings, 2 replies; 20+ messages in thread
From: Christophe Lombard @ 2017-03-14 11:08 UTC (permalink / raw)
To: linuxppc-dev, fbarrat, imunsie, andrew.donnellan
The mm_struct corresponding to the current task is acquired each time
an interrupt is raised. So to simplify the code, we only get the
mm_struct when attaching an AFU context to the process.
The mm_count reference is increased to ensure that the mm_struct can't
be freed. The mm_struct will be released when the context is detached.
The reference (use count) on the struct mm is not kept to avoid a
circular dependency if the process mmaps its cxl mmio and forget to
unmap before exiting.
Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com>
---
drivers/misc/cxl/api.c | 17 ++++++++--
drivers/misc/cxl/context.c | 26 ++++++++++++++--
drivers/misc/cxl/cxl.h | 13 ++++++--
drivers/misc/cxl/fault.c | 77 +++++-----------------------------------------
drivers/misc/cxl/file.c | 15 +++++++--
5 files changed, 68 insertions(+), 80 deletions(-)
diff --git a/drivers/misc/cxl/api.c b/drivers/misc/cxl/api.c
index bcc030e..1a138c8 100644
--- a/drivers/misc/cxl/api.c
+++ b/drivers/misc/cxl/api.c
@@ -14,6 +14,7 @@
#include <linux/msi.h>
#include <linux/module.h>
#include <linux/mount.h>
+#include <linux/sched/mm.h>
#include "cxl.h"
@@ -321,19 +322,29 @@ int cxl_start_context(struct cxl_context *ctx, u64 wed,
if (task) {
ctx->pid = get_task_pid(task, PIDTYPE_PID);
- ctx->glpid = get_task_pid(task->group_leader, PIDTYPE_PID);
kernel = false;
ctx->real_mode = false;
+
+ /* acquire a reference to the task's mm */
+ ctx->mm = get_task_mm(current);
+
+ /* ensure this mm_struct can't be freed */
+ cxl_context_mm_count_get(ctx);
+
+ /* decrement the use count */
+ if (ctx->mm)
+ mmput(ctx->mm);
}
cxl_ctx_get();
if ((rc = cxl_ops->attach_process(ctx, kernel, wed, 0))) {
- put_pid(ctx->glpid);
put_pid(ctx->pid);
- ctx->glpid = ctx->pid = NULL;
+ ctx->pid = NULL;
cxl_adapter_context_put(ctx->afu->adapter);
cxl_ctx_put();
+ if (task)
+ cxl_context_mm_count_put(ctx);
goto out;
}
diff --git a/drivers/misc/cxl/context.c b/drivers/misc/cxl/context.c
index 062bf6c..ed0a447 100644
--- a/drivers/misc/cxl/context.c
+++ b/drivers/misc/cxl/context.c
@@ -17,6 +17,7 @@
#include <linux/debugfs.h>
#include <linux/slab.h>
#include <linux/idr.h>
+#include <linux/sched/mm.h>
#include <asm/cputable.h>
#include <asm/current.h>
#include <asm/copro.h>
@@ -41,7 +42,7 @@ int cxl_context_init(struct cxl_context *ctx, struct cxl_afu *afu, bool master)
spin_lock_init(&ctx->sste_lock);
ctx->afu = afu;
ctx->master = master;
- ctx->pid = ctx->glpid = NULL; /* Set in start work ioctl */
+ ctx->pid = NULL; /* Set in start work ioctl */
mutex_init(&ctx->mapping_lock);
ctx->mapping = NULL;
@@ -242,12 +243,15 @@ int __detach_context(struct cxl_context *ctx)
/* release the reference to the group leader and mm handling pid */
put_pid(ctx->pid);
- put_pid(ctx->glpid);
cxl_ctx_put();
/* Decrease the attached context count on the adapter */
cxl_adapter_context_put(ctx->afu->adapter);
+
+ /* Decrease the mm count on the context */
+ cxl_context_mm_count_put(ctx);
+
return 0;
}
@@ -325,3 +329,21 @@ void cxl_context_free(struct cxl_context *ctx)
mutex_unlock(&ctx->afu->contexts_lock);
call_rcu(&ctx->rcu, reclaim_ctx);
}
+
+void cxl_context_mm_count_get(struct cxl_context *ctx)
+{
+ if (ctx->mm)
+ atomic_inc(&ctx->mm->mm_count);
+}
+
+void cxl_context_mm_count_put(struct cxl_context *ctx)
+{
+ if (ctx->mm)
+ mmdrop(ctx->mm);
+}
+
+void cxl_context_mm_users_get(struct cxl_context *ctx)
+{
+ if (ctx->mm)
+ atomic_inc(&ctx->mm->mm_users);
+}
diff --git a/drivers/misc/cxl/cxl.h b/drivers/misc/cxl/cxl.h
index 79e60ec..4d1b704 100644
--- a/drivers/misc/cxl/cxl.h
+++ b/drivers/misc/cxl/cxl.h
@@ -482,8 +482,6 @@ struct cxl_context {
unsigned int sst_size, sst_lru;
wait_queue_head_t wq;
- /* pid of the group leader associated with the pid */
- struct pid *glpid;
/* use mm context associated with this pid for ds faults */
struct pid *pid;
spinlock_t lock; /* Protects pending_irq_mask, pending_fault and fault_addr */
@@ -551,6 +549,8 @@ struct cxl_context {
* CX4 only:
*/
struct list_head extra_irq_contexts;
+
+ struct mm_struct *mm;
};
struct cxl_service_layer_ops {
@@ -1024,4 +1024,13 @@ int cxl_adapter_context_lock(struct cxl *adapter);
/* Unlock the contexts-lock if taken. Warn and force unlock otherwise */
void cxl_adapter_context_unlock(struct cxl *adapter);
+/* Increases the reference count to "struct mm_struct" */
+void cxl_context_mm_count_get(struct cxl_context *ctx);
+
+/* Decrements the reference count to "struct mm_struct" */
+void cxl_context_mm_count_put(struct cxl_context *ctx);
+
+/* Increases the reference users to "struct mm_struct" */
+void cxl_context_mm_users_get(struct cxl_context *ctx);
+
#endif
diff --git a/drivers/misc/cxl/fault.c b/drivers/misc/cxl/fault.c
index 2fa015c..14a5bfa 100644
--- a/drivers/misc/cxl/fault.c
+++ b/drivers/misc/cxl/fault.c
@@ -170,81 +170,19 @@ static void cxl_handle_page_fault(struct cxl_context *ctx,
}
/*
- * Returns the mm_struct corresponding to the context ctx via ctx->pid
- * In case the task has exited we use the task group leader accessible
- * via ctx->glpid to find the next task in the thread group that has a
- * valid mm_struct associated with it. If a task with valid mm_struct
- * is found the ctx->pid is updated to use the task struct for subsequent
- * translations. In case no valid mm_struct is found in the task group to
- * service the fault a NULL is returned.
+ * Returns the mm_struct corresponding to the context ctx.
+ * mm_users == 0, the context may be in the process of being closed.
*/
static struct mm_struct *get_mem_context(struct cxl_context *ctx)
{
- struct task_struct *task = NULL;
- struct mm_struct *mm = NULL;
- struct pid *old_pid = ctx->pid;
-
- if (old_pid == NULL) {
- pr_warn("%s: Invalid context for pe=%d\n",
- __func__, ctx->pe);
+ if (ctx->mm == NULL)
return NULL;
- }
-
- task = get_pid_task(old_pid, PIDTYPE_PID);
-
- /*
- * pid_alive may look racy but this saves us from costly
- * get_task_mm when the task is a zombie. In worst case
- * we may think a task is alive, which is about to die
- * but get_task_mm will return NULL.
- */
- if (task != NULL && pid_alive(task))
- mm = get_task_mm(task);
- /* release the task struct that was taken earlier */
- if (task)
- put_task_struct(task);
- else
- pr_devel("%s: Context owning pid=%i for pe=%i dead\n",
- __func__, pid_nr(old_pid), ctx->pe);
-
- /*
- * If we couldn't find the mm context then use the group
- * leader to iterate over the task group and find a task
- * that gives us mm_struct.
- */
- if (unlikely(mm == NULL && ctx->glpid != NULL)) {
-
- rcu_read_lock();
- task = pid_task(ctx->glpid, PIDTYPE_PID);
- if (task)
- do {
- mm = get_task_mm(task);
- if (mm) {
- ctx->pid = get_task_pid(task,
- PIDTYPE_PID);
- break;
- }
- task = next_thread(task);
- } while (task && !thread_group_leader(task));
- rcu_read_unlock();
-
- /* check if we switched pid */
- if (ctx->pid != old_pid) {
- if (mm)
- pr_devel("%s:pe=%i switch pid %i->%i\n",
- __func__, ctx->pe, pid_nr(old_pid),
- pid_nr(ctx->pid));
- else
- pr_devel("%s:Cannot find mm for pid=%i\n",
- __func__, pid_nr(old_pid));
-
- /* drop the reference to older pid */
- put_pid(old_pid);
- }
- }
+ if (atomic_read(&ctx->mm->mm_users) == 0)
+ return NULL;
- return mm;
+ cxl_context_mm_users_get(ctx);
+ return ctx->mm;
}
@@ -282,7 +220,6 @@ void cxl_handle_fault(struct work_struct *fault_work)
if (!ctx->kernel) {
mm = get_mem_context(ctx);
- /* indicates all the thread in task group have exited */
if (mm == NULL) {
pr_devel("%s: unable to get mm for pe=%d pid=%i\n",
__func__, ctx->pe, pid_nr(ctx->pid));
diff --git a/drivers/misc/cxl/file.c b/drivers/misc/cxl/file.c
index e7139c7..17b433f 100644
--- a/drivers/misc/cxl/file.c
+++ b/drivers/misc/cxl/file.c
@@ -18,6 +18,7 @@
#include <linux/fs.h>
#include <linux/mm.h>
#include <linux/slab.h>
+#include <linux/sched/mm.h>
#include <asm/cputable.h>
#include <asm/current.h>
#include <asm/copro.h>
@@ -216,8 +217,16 @@ static long afu_ioctl_start_work(struct cxl_context *ctx,
* process is still accessible.
*/
ctx->pid = get_task_pid(current, PIDTYPE_PID);
- ctx->glpid = get_task_pid(current->group_leader, PIDTYPE_PID);
+ /* acquire a reference to the task's mm */
+ ctx->mm = get_task_mm(current);
+
+ /* ensure this mm_struct can't be freed */
+ cxl_context_mm_count_get(ctx);
+
+ /* decrement the use count */
+ if (ctx->mm)
+ mmput(ctx->mm);
trace_cxl_attach(ctx, work.work_element_descriptor, work.num_interrupts, amr);
@@ -225,9 +234,9 @@ static long afu_ioctl_start_work(struct cxl_context *ctx,
amr))) {
afu_release_irqs(ctx, ctx);
cxl_adapter_context_put(ctx->afu->adapter);
- put_pid(ctx->glpid);
put_pid(ctx->pid);
- ctx->glpid = ctx->pid = NULL;
+ ctx->pid = NULL;
+ cxl_context_mm_count_put(ctx);
goto out;
}
--
2.7.4
^ permalink raw reply related [flat|nested] 20+ messages in thread* Re: [PATCH V2 2/6] cxl: Keep track of mm struct associated with a context
2017-03-14 11:08 ` [PATCH V2 2/6] cxl: Keep track of mm struct associated with a context Christophe Lombard
@ 2017-03-20 14:27 ` Frederic Barrat
2017-03-21 11:28 ` Frederic Barrat
1 sibling, 0 replies; 20+ messages in thread
From: Frederic Barrat @ 2017-03-20 14:27 UTC (permalink / raw)
To: Christophe Lombard, linuxppc-dev, imunsie, andrew.donnellan
Le 14/03/2017 à 12:08, Christophe Lombard a écrit :
> The mm_struct corresponding to the current task is acquired each time
> an interrupt is raised. So to simplify the code, we only get the
> mm_struct when attaching an AFU context to the process.
> The mm_count reference is increased to ensure that the mm_struct can't
> be freed. The mm_struct will be released when the context is detached.
> The reference (use count) on the struct mm is not kept to avoid a
> circular dependency if the process mmaps its cxl mmio and forget to
> unmap before exiting.
I think it should be:
A reference on mm_users is not kept to avoid...
>
> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com>
> ---
> drivers/misc/cxl/api.c | 17 ++++++++--
> drivers/misc/cxl/context.c | 26 ++++++++++++++--
> drivers/misc/cxl/cxl.h | 13 ++++++--
> drivers/misc/cxl/fault.c | 77 +++++-----------------------------------------
> drivers/misc/cxl/file.c | 15 +++++++--
> 5 files changed, 68 insertions(+), 80 deletions(-)
>
> diff --git a/drivers/misc/cxl/api.c b/drivers/misc/cxl/api.c
> index bcc030e..1a138c8 100644
> --- a/drivers/misc/cxl/api.c
> +++ b/drivers/misc/cxl/api.c
> @@ -14,6 +14,7 @@
> #include <linux/msi.h>
> #include <linux/module.h>
> #include <linux/mount.h>
> +#include <linux/sched/mm.h>
>
> #include "cxl.h"
>
> @@ -321,19 +322,29 @@ int cxl_start_context(struct cxl_context *ctx, u64 wed,
>
> if (task) {
> ctx->pid = get_task_pid(task, PIDTYPE_PID);
> - ctx->glpid = get_task_pid(task->group_leader, PIDTYPE_PID);
> kernel = false;
> ctx->real_mode = false;
> +
> + /* acquire a reference to the task's mm */
> + ctx->mm = get_task_mm(current);
> +
> + /* ensure this mm_struct can't be freed */
> + cxl_context_mm_count_get(ctx);
> +
> + /* decrement the use count */
> + if (ctx->mm)
> + mmput(ctx->mm);
> }
>
> cxl_ctx_get();
>
> if ((rc = cxl_ops->attach_process(ctx, kernel, wed, 0))) {
> - put_pid(ctx->glpid);
> put_pid(ctx->pid);
> - ctx->glpid = ctx->pid = NULL;
> + ctx->pid = NULL;
> cxl_adapter_context_put(ctx->afu->adapter);
> cxl_ctx_put();
> + if (task)
> + cxl_context_mm_count_put(ctx);
> goto out;
> }
>
> diff --git a/drivers/misc/cxl/context.c b/drivers/misc/cxl/context.c
> index 062bf6c..ed0a447 100644
> --- a/drivers/misc/cxl/context.c
> +++ b/drivers/misc/cxl/context.c
> @@ -17,6 +17,7 @@
> #include <linux/debugfs.h>
> #include <linux/slab.h>
> #include <linux/idr.h>
> +#include <linux/sched/mm.h>
> #include <asm/cputable.h>
> #include <asm/current.h>
> #include <asm/copro.h>
> @@ -41,7 +42,7 @@ int cxl_context_init(struct cxl_context *ctx, struct cxl_afu *afu, bool master)
> spin_lock_init(&ctx->sste_lock);
> ctx->afu = afu;
> ctx->master = master;
> - ctx->pid = ctx->glpid = NULL; /* Set in start work ioctl */
> + ctx->pid = NULL; /* Set in start work ioctl */
> mutex_init(&ctx->mapping_lock);
> ctx->mapping = NULL;
>
> @@ -242,12 +243,15 @@ int __detach_context(struct cxl_context *ctx)
>
> /* release the reference to the group leader and mm handling pid */
> put_pid(ctx->pid);
> - put_pid(ctx->glpid);
>
> cxl_ctx_put();
>
> /* Decrease the attached context count on the adapter */
> cxl_adapter_context_put(ctx->afu->adapter);
> +
> + /* Decrease the mm count on the context */
> + cxl_context_mm_count_put(ctx);
> +
> return 0;
> }
>
> @@ -325,3 +329,21 @@ void cxl_context_free(struct cxl_context *ctx)
> mutex_unlock(&ctx->afu->contexts_lock);
> call_rcu(&ctx->rcu, reclaim_ctx);
> }
> +
> +void cxl_context_mm_count_get(struct cxl_context *ctx)
> +{
> + if (ctx->mm)
> + atomic_inc(&ctx->mm->mm_count);
> +}
> +
> +void cxl_context_mm_count_put(struct cxl_context *ctx)
> +{
> + if (ctx->mm)
> + mmdrop(ctx->mm);
> +}
> +
> +void cxl_context_mm_users_get(struct cxl_context *ctx)
> +{
> + if (ctx->mm)
> + atomic_inc(&ctx->mm->mm_users);
> +}
> diff --git a/drivers/misc/cxl/cxl.h b/drivers/misc/cxl/cxl.h
> index 79e60ec..4d1b704 100644
> --- a/drivers/misc/cxl/cxl.h
> +++ b/drivers/misc/cxl/cxl.h
> @@ -482,8 +482,6 @@ struct cxl_context {
> unsigned int sst_size, sst_lru;
>
> wait_queue_head_t wq;
> - /* pid of the group leader associated with the pid */
> - struct pid *glpid;
> /* use mm context associated with this pid for ds faults */
> struct pid *pid;
> spinlock_t lock; /* Protects pending_irq_mask, pending_fault and fault_addr */
> @@ -551,6 +549,8 @@ struct cxl_context {
> * CX4 only:
> */
> struct list_head extra_irq_contexts;
> +
> + struct mm_struct *mm;
> };
>
> struct cxl_service_layer_ops {
> @@ -1024,4 +1024,13 @@ int cxl_adapter_context_lock(struct cxl *adapter);
> /* Unlock the contexts-lock if taken. Warn and force unlock otherwise */
> void cxl_adapter_context_unlock(struct cxl *adapter);
>
> +/* Increases the reference count to "struct mm_struct" */
> +void cxl_context_mm_count_get(struct cxl_context *ctx);
> +
> +/* Decrements the reference count to "struct mm_struct" */
> +void cxl_context_mm_count_put(struct cxl_context *ctx);
> +
> +/* Increases the reference users to "struct mm_struct" */
> +void cxl_context_mm_users_get(struct cxl_context *ctx);
> +
> #endif
> diff --git a/drivers/misc/cxl/fault.c b/drivers/misc/cxl/fault.c
> index 2fa015c..14a5bfa 100644
> --- a/drivers/misc/cxl/fault.c
> +++ b/drivers/misc/cxl/fault.c
> @@ -170,81 +170,19 @@ static void cxl_handle_page_fault(struct cxl_context *ctx,
> }
>
> /*
> - * Returns the mm_struct corresponding to the context ctx via ctx->pid
> - * In case the task has exited we use the task group leader accessible
> - * via ctx->glpid to find the next task in the thread group that has a
> - * valid mm_struct associated with it. If a task with valid mm_struct
> - * is found the ctx->pid is updated to use the task struct for subsequent
> - * translations. In case no valid mm_struct is found in the task group to
> - * service the fault a NULL is returned.
> + * Returns the mm_struct corresponding to the context ctx.
> + * mm_users == 0, the context may be in the process of being closed.
> */
> static struct mm_struct *get_mem_context(struct cxl_context *ctx)
> {
> - struct task_struct *task = NULL;
> - struct mm_struct *mm = NULL;
> - struct pid *old_pid = ctx->pid;
> -
> - if (old_pid == NULL) {
> - pr_warn("%s: Invalid context for pe=%d\n",
> - __func__, ctx->pe);
> + if (ctx->mm == NULL)
> return NULL;
> - }
> -
> - task = get_pid_task(old_pid, PIDTYPE_PID);
> -
> - /*
> - * pid_alive may look racy but this saves us from costly
> - * get_task_mm when the task is a zombie. In worst case
> - * we may think a task is alive, which is about to die
> - * but get_task_mm will return NULL.
> - */
> - if (task != NULL && pid_alive(task))
> - mm = get_task_mm(task);
>
> - /* release the task struct that was taken earlier */
> - if (task)
> - put_task_struct(task);
> - else
> - pr_devel("%s: Context owning pid=%i for pe=%i dead\n",
> - __func__, pid_nr(old_pid), ctx->pe);
> -
> - /*
> - * If we couldn't find the mm context then use the group
> - * leader to iterate over the task group and find a task
> - * that gives us mm_struct.
> - */
> - if (unlikely(mm == NULL && ctx->glpid != NULL)) {
> -
> - rcu_read_lock();
> - task = pid_task(ctx->glpid, PIDTYPE_PID);
> - if (task)
> - do {
> - mm = get_task_mm(task);
> - if (mm) {
> - ctx->pid = get_task_pid(task,
> - PIDTYPE_PID);
> - break;
> - }
> - task = next_thread(task);
> - } while (task && !thread_group_leader(task));
> - rcu_read_unlock();
> -
> - /* check if we switched pid */
> - if (ctx->pid != old_pid) {
> - if (mm)
> - pr_devel("%s:pe=%i switch pid %i->%i\n",
> - __func__, ctx->pe, pid_nr(old_pid),
> - pid_nr(ctx->pid));
> - else
> - pr_devel("%s:Cannot find mm for pid=%i\n",
> - __func__, pid_nr(old_pid));
> -
> - /* drop the reference to older pid */
> - put_pid(old_pid);
> - }
> - }
> + if (atomic_read(&ctx->mm->mm_users) == 0)
> + return NULL;
>
> - return mm;
> + cxl_context_mm_users_get(ctx);
> + return ctx->mm;
It should be done atomically:
if (!atomic_inc_not_zero(&ctx->mm->mm_users))
return ctx->mm;
return NULL;
in which case, we don't need cxl_context_mm_users_get() any more.
Fred
> }
>
>
> @@ -282,7 +220,6 @@ void cxl_handle_fault(struct work_struct *fault_work)
> if (!ctx->kernel) {
>
> mm = get_mem_context(ctx);
> - /* indicates all the thread in task group have exited */
> if (mm == NULL) {
> pr_devel("%s: unable to get mm for pe=%d pid=%i\n",
> __func__, ctx->pe, pid_nr(ctx->pid));
> diff --git a/drivers/misc/cxl/file.c b/drivers/misc/cxl/file.c
> index e7139c7..17b433f 100644
> --- a/drivers/misc/cxl/file.c
> +++ b/drivers/misc/cxl/file.c
> @@ -18,6 +18,7 @@
> #include <linux/fs.h>
> #include <linux/mm.h>
> #include <linux/slab.h>
> +#include <linux/sched/mm.h>
> #include <asm/cputable.h>
> #include <asm/current.h>
> #include <asm/copro.h>
> @@ -216,8 +217,16 @@ static long afu_ioctl_start_work(struct cxl_context *ctx,
> * process is still accessible.
> */
> ctx->pid = get_task_pid(current, PIDTYPE_PID);
> - ctx->glpid = get_task_pid(current->group_leader, PIDTYPE_PID);
>
> + /* acquire a reference to the task's mm */
> + ctx->mm = get_task_mm(current);
> +
> + /* ensure this mm_struct can't be freed */
> + cxl_context_mm_count_get(ctx);
> +
> + /* decrement the use count */
> + if (ctx->mm)
> + mmput(ctx->mm);
>
> trace_cxl_attach(ctx, work.work_element_descriptor, work.num_interrupts, amr);
>
> @@ -225,9 +234,9 @@ static long afu_ioctl_start_work(struct cxl_context *ctx,
> amr))) {
> afu_release_irqs(ctx, ctx);
> cxl_adapter_context_put(ctx->afu->adapter);
> - put_pid(ctx->glpid);
> put_pid(ctx->pid);
> - ctx->glpid = ctx->pid = NULL;
> + ctx->pid = NULL;
> + cxl_context_mm_count_put(ctx);
> goto out;
> }
>
^ permalink raw reply [flat|nested] 20+ messages in thread* Re: [PATCH V2 2/6] cxl: Keep track of mm struct associated with a context
2017-03-14 11:08 ` [PATCH V2 2/6] cxl: Keep track of mm struct associated with a context Christophe Lombard
2017-03-20 14:27 ` Frederic Barrat
@ 2017-03-21 11:28 ` Frederic Barrat
1 sibling, 0 replies; 20+ messages in thread
From: Frederic Barrat @ 2017-03-21 11:28 UTC (permalink / raw)
To: Christophe Lombard, linuxppc-dev, imunsie, andrew.donnellan
Another thought about that patch. Now that we keep track of the mm
associated to a context, I think we can simplify slightly the function
_cxl_slbia() in main.c, where we look for the mm based on the pid. We
now have the information readily available.
Fred
Le 14/03/2017 à 12:08, Christophe Lombard a écrit :
> The mm_struct corresponding to the current task is acquired each time
> an interrupt is raised. So to simplify the code, we only get the
> mm_struct when attaching an AFU context to the process.
> The mm_count reference is increased to ensure that the mm_struct can't
> be freed. The mm_struct will be released when the context is detached.
> The reference (use count) on the struct mm is not kept to avoid a
> circular dependency if the process mmaps its cxl mmio and forget to
> unmap before exiting.
>
> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com>
> ---
> drivers/misc/cxl/api.c | 17 ++++++++--
> drivers/misc/cxl/context.c | 26 ++++++++++++++--
> drivers/misc/cxl/cxl.h | 13 ++++++--
> drivers/misc/cxl/fault.c | 77 +++++-----------------------------------------
> drivers/misc/cxl/file.c | 15 +++++++--
> 5 files changed, 68 insertions(+), 80 deletions(-)
>
> diff --git a/drivers/misc/cxl/api.c b/drivers/misc/cxl/api.c
> index bcc030e..1a138c8 100644
> --- a/drivers/misc/cxl/api.c
> +++ b/drivers/misc/cxl/api.c
> @@ -14,6 +14,7 @@
> #include <linux/msi.h>
> #include <linux/module.h>
> #include <linux/mount.h>
> +#include <linux/sched/mm.h>
>
> #include "cxl.h"
>
> @@ -321,19 +322,29 @@ int cxl_start_context(struct cxl_context *ctx, u64 wed,
>
> if (task) {
> ctx->pid = get_task_pid(task, PIDTYPE_PID);
> - ctx->glpid = get_task_pid(task->group_leader, PIDTYPE_PID);
> kernel = false;
> ctx->real_mode = false;
> +
> + /* acquire a reference to the task's mm */
> + ctx->mm = get_task_mm(current);
> +
> + /* ensure this mm_struct can't be freed */
> + cxl_context_mm_count_get(ctx);
> +
> + /* decrement the use count */
> + if (ctx->mm)
> + mmput(ctx->mm);
> }
>
> cxl_ctx_get();
>
> if ((rc = cxl_ops->attach_process(ctx, kernel, wed, 0))) {
> - put_pid(ctx->glpid);
> put_pid(ctx->pid);
> - ctx->glpid = ctx->pid = NULL;
> + ctx->pid = NULL;
> cxl_adapter_context_put(ctx->afu->adapter);
> cxl_ctx_put();
> + if (task)
> + cxl_context_mm_count_put(ctx);
> goto out;
> }
>
> diff --git a/drivers/misc/cxl/context.c b/drivers/misc/cxl/context.c
> index 062bf6c..ed0a447 100644
> --- a/drivers/misc/cxl/context.c
> +++ b/drivers/misc/cxl/context.c
> @@ -17,6 +17,7 @@
> #include <linux/debugfs.h>
> #include <linux/slab.h>
> #include <linux/idr.h>
> +#include <linux/sched/mm.h>
> #include <asm/cputable.h>
> #include <asm/current.h>
> #include <asm/copro.h>
> @@ -41,7 +42,7 @@ int cxl_context_init(struct cxl_context *ctx, struct cxl_afu *afu, bool master)
> spin_lock_init(&ctx->sste_lock);
> ctx->afu = afu;
> ctx->master = master;
> - ctx->pid = ctx->glpid = NULL; /* Set in start work ioctl */
> + ctx->pid = NULL; /* Set in start work ioctl */
> mutex_init(&ctx->mapping_lock);
> ctx->mapping = NULL;
>
> @@ -242,12 +243,15 @@ int __detach_context(struct cxl_context *ctx)
>
> /* release the reference to the group leader and mm handling pid */
> put_pid(ctx->pid);
> - put_pid(ctx->glpid);
>
> cxl_ctx_put();
>
> /* Decrease the attached context count on the adapter */
> cxl_adapter_context_put(ctx->afu->adapter);
> +
> + /* Decrease the mm count on the context */
> + cxl_context_mm_count_put(ctx);
> +
> return 0;
> }
>
> @@ -325,3 +329,21 @@ void cxl_context_free(struct cxl_context *ctx)
> mutex_unlock(&ctx->afu->contexts_lock);
> call_rcu(&ctx->rcu, reclaim_ctx);
> }
> +
> +void cxl_context_mm_count_get(struct cxl_context *ctx)
> +{
> + if (ctx->mm)
> + atomic_inc(&ctx->mm->mm_count);
> +}
> +
> +void cxl_context_mm_count_put(struct cxl_context *ctx)
> +{
> + if (ctx->mm)
> + mmdrop(ctx->mm);
> +}
> +
> +void cxl_context_mm_users_get(struct cxl_context *ctx)
> +{
> + if (ctx->mm)
> + atomic_inc(&ctx->mm->mm_users);
> +}
> diff --git a/drivers/misc/cxl/cxl.h b/drivers/misc/cxl/cxl.h
> index 79e60ec..4d1b704 100644
> --- a/drivers/misc/cxl/cxl.h
> +++ b/drivers/misc/cxl/cxl.h
> @@ -482,8 +482,6 @@ struct cxl_context {
> unsigned int sst_size, sst_lru;
>
> wait_queue_head_t wq;
> - /* pid of the group leader associated with the pid */
> - struct pid *glpid;
> /* use mm context associated with this pid for ds faults */
> struct pid *pid;
> spinlock_t lock; /* Protects pending_irq_mask, pending_fault and fault_addr */
> @@ -551,6 +549,8 @@ struct cxl_context {
> * CX4 only:
> */
> struct list_head extra_irq_contexts;
> +
> + struct mm_struct *mm;
> };
>
> struct cxl_service_layer_ops {
> @@ -1024,4 +1024,13 @@ int cxl_adapter_context_lock(struct cxl *adapter);
> /* Unlock the contexts-lock if taken. Warn and force unlock otherwise */
> void cxl_adapter_context_unlock(struct cxl *adapter);
>
> +/* Increases the reference count to "struct mm_struct" */
> +void cxl_context_mm_count_get(struct cxl_context *ctx);
> +
> +/* Decrements the reference count to "struct mm_struct" */
> +void cxl_context_mm_count_put(struct cxl_context *ctx);
> +
> +/* Increases the reference users to "struct mm_struct" */
> +void cxl_context_mm_users_get(struct cxl_context *ctx);
> +
> #endif
> diff --git a/drivers/misc/cxl/fault.c b/drivers/misc/cxl/fault.c
> index 2fa015c..14a5bfa 100644
> --- a/drivers/misc/cxl/fault.c
> +++ b/drivers/misc/cxl/fault.c
> @@ -170,81 +170,19 @@ static void cxl_handle_page_fault(struct cxl_context *ctx,
> }
>
> /*
> - * Returns the mm_struct corresponding to the context ctx via ctx->pid
> - * In case the task has exited we use the task group leader accessible
> - * via ctx->glpid to find the next task in the thread group that has a
> - * valid mm_struct associated with it. If a task with valid mm_struct
> - * is found the ctx->pid is updated to use the task struct for subsequent
> - * translations. In case no valid mm_struct is found in the task group to
> - * service the fault a NULL is returned.
> + * Returns the mm_struct corresponding to the context ctx.
> + * mm_users == 0, the context may be in the process of being closed.
> */
> static struct mm_struct *get_mem_context(struct cxl_context *ctx)
> {
> - struct task_struct *task = NULL;
> - struct mm_struct *mm = NULL;
> - struct pid *old_pid = ctx->pid;
> -
> - if (old_pid == NULL) {
> - pr_warn("%s: Invalid context for pe=%d\n",
> - __func__, ctx->pe);
> + if (ctx->mm == NULL)
> return NULL;
> - }
> -
> - task = get_pid_task(old_pid, PIDTYPE_PID);
> -
> - /*
> - * pid_alive may look racy but this saves us from costly
> - * get_task_mm when the task is a zombie. In worst case
> - * we may think a task is alive, which is about to die
> - * but get_task_mm will return NULL.
> - */
> - if (task != NULL && pid_alive(task))
> - mm = get_task_mm(task);
>
> - /* release the task struct that was taken earlier */
> - if (task)
> - put_task_struct(task);
> - else
> - pr_devel("%s: Context owning pid=%i for pe=%i dead\n",
> - __func__, pid_nr(old_pid), ctx->pe);
> -
> - /*
> - * If we couldn't find the mm context then use the group
> - * leader to iterate over the task group and find a task
> - * that gives us mm_struct.
> - */
> - if (unlikely(mm == NULL && ctx->glpid != NULL)) {
> -
> - rcu_read_lock();
> - task = pid_task(ctx->glpid, PIDTYPE_PID);
> - if (task)
> - do {
> - mm = get_task_mm(task);
> - if (mm) {
> - ctx->pid = get_task_pid(task,
> - PIDTYPE_PID);
> - break;
> - }
> - task = next_thread(task);
> - } while (task && !thread_group_leader(task));
> - rcu_read_unlock();
> -
> - /* check if we switched pid */
> - if (ctx->pid != old_pid) {
> - if (mm)
> - pr_devel("%s:pe=%i switch pid %i->%i\n",
> - __func__, ctx->pe, pid_nr(old_pid),
> - pid_nr(ctx->pid));
> - else
> - pr_devel("%s:Cannot find mm for pid=%i\n",
> - __func__, pid_nr(old_pid));
> -
> - /* drop the reference to older pid */
> - put_pid(old_pid);
> - }
> - }
> + if (atomic_read(&ctx->mm->mm_users) == 0)
> + return NULL;
>
> - return mm;
> + cxl_context_mm_users_get(ctx);
> + return ctx->mm;
> }
>
>
> @@ -282,7 +220,6 @@ void cxl_handle_fault(struct work_struct *fault_work)
> if (!ctx->kernel) {
>
> mm = get_mem_context(ctx);
> - /* indicates all the thread in task group have exited */
> if (mm == NULL) {
> pr_devel("%s: unable to get mm for pe=%d pid=%i\n",
> __func__, ctx->pe, pid_nr(ctx->pid));
> diff --git a/drivers/misc/cxl/file.c b/drivers/misc/cxl/file.c
> index e7139c7..17b433f 100644
> --- a/drivers/misc/cxl/file.c
> +++ b/drivers/misc/cxl/file.c
> @@ -18,6 +18,7 @@
> #include <linux/fs.h>
> #include <linux/mm.h>
> #include <linux/slab.h>
> +#include <linux/sched/mm.h>
> #include <asm/cputable.h>
> #include <asm/current.h>
> #include <asm/copro.h>
> @@ -216,8 +217,16 @@ static long afu_ioctl_start_work(struct cxl_context *ctx,
> * process is still accessible.
> */
> ctx->pid = get_task_pid(current, PIDTYPE_PID);
> - ctx->glpid = get_task_pid(current->group_leader, PIDTYPE_PID);
>
> + /* acquire a reference to the task's mm */
> + ctx->mm = get_task_mm(current);
> +
> + /* ensure this mm_struct can't be freed */
> + cxl_context_mm_count_get(ctx);
> +
> + /* decrement the use count */
> + if (ctx->mm)
> + mmput(ctx->mm);
>
> trace_cxl_attach(ctx, work.work_element_descriptor, work.num_interrupts, amr);
>
> @@ -225,9 +234,9 @@ static long afu_ioctl_start_work(struct cxl_context *ctx,
> amr))) {
> afu_release_irqs(ctx, ctx);
> cxl_adapter_context_put(ctx->afu->adapter);
> - put_pid(ctx->glpid);
> put_pid(ctx->pid);
> - ctx->glpid = ctx->pid = NULL;
> + ctx->pid = NULL;
> + cxl_context_mm_count_put(ctx);
> goto out;
> }
>
^ permalink raw reply [flat|nested] 20+ messages in thread
* [PATCH V2 3/6] cxl: Update implementation service layer
2017-03-14 11:08 [PATCH V2 0/6] cxl: Add support for Coherent Accelerator Interface Architecture 2.0 Christophe Lombard
2017-03-14 11:08 ` [PATCH V2 1/6] cxl: Remove unused values in bare-metal environment Christophe Lombard
2017-03-14 11:08 ` [PATCH V2 2/6] cxl: Keep track of mm struct associated with a context Christophe Lombard
@ 2017-03-14 11:08 ` Christophe Lombard
2017-03-20 16:17 ` Frederic Barrat
2017-03-14 11:08 ` [PATCH V2 4/6] cxl: Rename some psl8 specific functions Christophe Lombard
` (3 subsequent siblings)
6 siblings, 1 reply; 20+ messages in thread
From: Christophe Lombard @ 2017-03-14 11:08 UTC (permalink / raw)
To: linuxppc-dev, fbarrat, imunsie, andrew.donnellan
The service layer API (in cxl.h) lists some low-level functions whose
implementation is different on PSL8, PSL9 and XSL:
- Init implementation for the adapter and the afu.
- Invalidate TLB/SLB.
- Attach process for dedicated/directed models.
- Handle psl interrupts.
- Debug registers for the adapter and the afu.
- Traces.
Each environment implements its own functions, and the common code uses
them through function pointers, defined in cxl_service_layer_ops.
Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com>
---
drivers/misc/cxl/cxl.h | 34 ++++++++++++++++++++++---------
drivers/misc/cxl/debugfs.c | 16 +++++++--------
drivers/misc/cxl/guest.c | 2 +-
drivers/misc/cxl/irq.c | 2 +-
drivers/misc/cxl/native.c | 50 +++++++++++++++++++++++++++-------------------
drivers/misc/cxl/pci.c | 47 +++++++++++++++++++++++++++++--------------
6 files changed, 97 insertions(+), 54 deletions(-)
diff --git a/drivers/misc/cxl/cxl.h b/drivers/misc/cxl/cxl.h
index 4d1b704..3e03a66 100644
--- a/drivers/misc/cxl/cxl.h
+++ b/drivers/misc/cxl/cxl.h
@@ -553,13 +553,23 @@ struct cxl_context {
struct mm_struct *mm;
};
+struct cxl_irq_info;
+
struct cxl_service_layer_ops {
int (*adapter_regs_init)(struct cxl *adapter, struct pci_dev *dev);
+ int (*invalidate_all)(struct cxl *adapter);
int (*afu_regs_init)(struct cxl_afu *afu);
+ int (*sanitise_afu_regs)(struct cxl_afu *afu);
int (*register_serr_irq)(struct cxl_afu *afu);
void (*release_serr_irq)(struct cxl_afu *afu);
- void (*debugfs_add_adapter_sl_regs)(struct cxl *adapter, struct dentry *dir);
- void (*debugfs_add_afu_sl_regs)(struct cxl_afu *afu, struct dentry *dir);
+ irqreturn_t (*handle_interrupt)(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info);
+ irqreturn_t (*fail_irq)(struct cxl_afu *afu, struct cxl_irq_info *irq_info);
+ int (*activate_dedicated_process)(struct cxl_afu *afu);
+ int (*attach_afu_directed)(struct cxl_context *ctx, u64 wed, u64 amr);
+ int (*attach_dedicated_process)(struct cxl_context *ctx, u64 wed, u64 amr);
+ void (*update_dedicated_ivtes)(struct cxl_context *ctx);
+ void (*debugfs_add_adapter_regs)(struct cxl *adapter, struct dentry *dir);
+ void (*debugfs_add_afu_regs)(struct cxl_afu *afu, struct dentry *dir);
void (*psl_irq_dump_registers)(struct cxl_context *ctx);
void (*err_irq_dump_registers)(struct cxl *adapter);
void (*debugfs_stop_trace)(struct cxl *adapter);
@@ -805,16 +815,21 @@ void afu_irq_name_free(struct cxl_context *ctx);
#ifdef CONFIG_DEBUG_FS
+int cxl_attach_afu_directed_psl(struct cxl_context *ctx, u64 wed, u64 amr);
+int cxl_activate_dedicated_process_psl(struct cxl_afu *afu);
+int cxl_attach_dedicated_process_psl(struct cxl_context *ctx, u64 wed, u64 amr);
+void cxl_update_dedicated_ivtes_psl(struct cxl_context *ctx);
+
int cxl_debugfs_init(void);
void cxl_debugfs_exit(void);
int cxl_debugfs_adapter_add(struct cxl *adapter);
void cxl_debugfs_adapter_remove(struct cxl *adapter);
int cxl_debugfs_afu_add(struct cxl_afu *afu);
void cxl_debugfs_afu_remove(struct cxl_afu *afu);
-void cxl_stop_trace(struct cxl *cxl);
-void cxl_debugfs_add_adapter_psl_regs(struct cxl *adapter, struct dentry *dir);
-void cxl_debugfs_add_adapter_xsl_regs(struct cxl *adapter, struct dentry *dir);
-void cxl_debugfs_add_afu_psl_regs(struct cxl_afu *afu, struct dentry *dir);
+void cxl_stop_trace_psl(struct cxl *cxl);
+void cxl_debugfs_add_adapter_regs_psl(struct cxl *adapter, struct dentry *dir);
+void cxl_debugfs_add_adapter_regs_xsl(struct cxl *adapter, struct dentry *dir);
+void cxl_debugfs_add_afu_regs_psl(struct cxl_afu *afu, struct dentry *dir);
#else /* CONFIG_DEBUG_FS */
@@ -916,19 +931,20 @@ struct cxl_irq_info {
};
void cxl_assign_psn_space(struct cxl_context *ctx);
-irqreturn_t cxl_irq(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info);
+int cxl_invalidate_all_psl(struct cxl *adapter);
+irqreturn_t cxl_irq_psl(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info);
+irqreturn_t cxl_fail_irq_psl(struct cxl_afu *afu, struct cxl_irq_info *irq_info);
int cxl_register_one_irq(struct cxl *adapter, irq_handler_t handler,
void *cookie, irq_hw_number_t *dest_hwirq,
unsigned int *dest_virq, const char *name);
int cxl_check_error(struct cxl_afu *afu);
int cxl_afu_slbia(struct cxl_afu *afu);
-int cxl_tlb_slb_invalidate(struct cxl *adapter);
int cxl_data_cache_flush(struct cxl *adapter);
int cxl_afu_disable(struct cxl_afu *afu);
int cxl_psl_purge(struct cxl_afu *afu);
-void cxl_native_psl_irq_dump_regs(struct cxl_context *ctx);
+void cxl_native_irq_dump_regs_psl(struct cxl_context *ctx);
void cxl_native_err_irq_dump_regs(struct cxl *adapter);
int cxl_pci_vphb_add(struct cxl_afu *afu);
void cxl_pci_vphb_remove(struct cxl_afu *afu);
diff --git a/drivers/misc/cxl/debugfs.c b/drivers/misc/cxl/debugfs.c
index 9c06ac8..4848ebf 100644
--- a/drivers/misc/cxl/debugfs.c
+++ b/drivers/misc/cxl/debugfs.c
@@ -15,7 +15,7 @@
static struct dentry *cxl_debugfs;
-void cxl_stop_trace(struct cxl *adapter)
+void cxl_stop_trace_psl(struct cxl *adapter)
{
int slice;
@@ -53,7 +53,7 @@ static struct dentry *debugfs_create_io_x64(const char *name, umode_t mode,
(void __force *)value, &fops_io_x64);
}
-void cxl_debugfs_add_adapter_psl_regs(struct cxl *adapter, struct dentry *dir)
+void cxl_debugfs_add_adapter_regs_psl(struct cxl *adapter, struct dentry *dir)
{
debugfs_create_io_x64("fir1", S_IRUSR, dir, _cxl_p1_addr(adapter, CXL_PSL_FIR1));
debugfs_create_io_x64("fir2", S_IRUSR, dir, _cxl_p1_addr(adapter, CXL_PSL_FIR2));
@@ -61,7 +61,7 @@ void cxl_debugfs_add_adapter_psl_regs(struct cxl *adapter, struct dentry *dir)
debugfs_create_io_x64("trace", S_IRUSR | S_IWUSR, dir, _cxl_p1_addr(adapter, CXL_PSL_TRACE));
}
-void cxl_debugfs_add_adapter_xsl_regs(struct cxl *adapter, struct dentry *dir)
+void cxl_debugfs_add_adapter_regs_xsl(struct cxl *adapter, struct dentry *dir)
{
debugfs_create_io_x64("fec", S_IRUSR, dir, _cxl_p1_addr(adapter, CXL_XSL_FEC));
}
@@ -82,8 +82,8 @@ int cxl_debugfs_adapter_add(struct cxl *adapter)
debugfs_create_io_x64("err_ivte", S_IRUSR, dir, _cxl_p1_addr(adapter, CXL_PSL_ErrIVTE));
- if (adapter->native->sl_ops->debugfs_add_adapter_sl_regs)
- adapter->native->sl_ops->debugfs_add_adapter_sl_regs(adapter, dir);
+ if (adapter->native->sl_ops->debugfs_add_adapter_regs)
+ adapter->native->sl_ops->debugfs_add_adapter_regs(adapter, dir);
return 0;
}
@@ -92,7 +92,7 @@ void cxl_debugfs_adapter_remove(struct cxl *adapter)
debugfs_remove_recursive(adapter->debugfs);
}
-void cxl_debugfs_add_afu_psl_regs(struct cxl_afu *afu, struct dentry *dir)
+void cxl_debugfs_add_afu_regs_psl(struct cxl_afu *afu, struct dentry *dir)
{
debugfs_create_io_x64("fir", S_IRUSR, dir, _cxl_p1n_addr(afu, CXL_PSL_FIR_SLICE_An));
debugfs_create_io_x64("serr", S_IRUSR, dir, _cxl_p1n_addr(afu, CXL_PSL_SERR_An));
@@ -121,8 +121,8 @@ int cxl_debugfs_afu_add(struct cxl_afu *afu)
debugfs_create_io_x64("sstp1", S_IRUSR, dir, _cxl_p2n_addr(afu, CXL_SSTP1_An));
debugfs_create_io_x64("err_status", S_IRUSR, dir, _cxl_p2n_addr(afu, CXL_PSL_ErrStat_An));
- if (afu->adapter->native->sl_ops->debugfs_add_afu_sl_regs)
- afu->adapter->native->sl_ops->debugfs_add_afu_sl_regs(afu, dir);
+ if (afu->adapter->native->sl_ops->debugfs_add_afu_regs)
+ afu->adapter->native->sl_ops->debugfs_add_afu_regs(afu, dir);
return 0;
}
diff --git a/drivers/misc/cxl/guest.c b/drivers/misc/cxl/guest.c
index e04bc4d..f6ba698 100644
--- a/drivers/misc/cxl/guest.c
+++ b/drivers/misc/cxl/guest.c
@@ -169,7 +169,7 @@ static irqreturn_t guest_psl_irq(int irq, void *data)
return IRQ_HANDLED;
}
- rc = cxl_irq(irq, ctx, &irq_info);
+ rc = cxl_irq_psl(irq, ctx, &irq_info);
return rc;
}
diff --git a/drivers/misc/cxl/irq.c b/drivers/misc/cxl/irq.c
index 1a402bb..2fa119e 100644
--- a/drivers/misc/cxl/irq.c
+++ b/drivers/misc/cxl/irq.c
@@ -34,7 +34,7 @@ static irqreturn_t schedule_cxl_fault(struct cxl_context *ctx, u64 dsisr, u64 da
return IRQ_HANDLED;
}
-irqreturn_t cxl_irq(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info)
+irqreturn_t cxl_irq_psl(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info)
{
u64 dsisr, dar;
diff --git a/drivers/misc/cxl/native.c b/drivers/misc/cxl/native.c
index 7257e8b..2d1b0c6 100644
--- a/drivers/misc/cxl/native.c
+++ b/drivers/misc/cxl/native.c
@@ -258,7 +258,7 @@ void cxl_release_spa(struct cxl_afu *afu)
}
}
-int cxl_tlb_slb_invalidate(struct cxl *adapter)
+int cxl_invalidate_all_psl(struct cxl *adapter)
{
unsigned long timeout = jiffies + (HZ * CXL_TIMEOUT);
@@ -578,7 +578,7 @@ static void update_ivtes_directed(struct cxl_context *ctx)
WARN_ON(add_process_element(ctx));
}
-static int attach_afu_directed(struct cxl_context *ctx, u64 wed, u64 amr)
+int cxl_attach_afu_directed_psl(struct cxl_context *ctx, u64 wed, u64 amr)
{
u32 pid;
int result;
@@ -671,7 +671,7 @@ static int deactivate_afu_directed(struct cxl_afu *afu)
return 0;
}
-static int activate_dedicated_process(struct cxl_afu *afu)
+int cxl_activate_dedicated_process_psl(struct cxl_afu *afu)
{
dev_info(&afu->dev, "Activating dedicated process mode\n");
@@ -694,7 +694,7 @@ static int activate_dedicated_process(struct cxl_afu *afu)
return cxl_chardev_d_afu_add(afu);
}
-static void update_ivtes_dedicated(struct cxl_context *ctx)
+void cxl_update_dedicated_ivtes_psl(struct cxl_context *ctx)
{
struct cxl_afu *afu = ctx->afu;
@@ -710,7 +710,7 @@ static void update_ivtes_dedicated(struct cxl_context *ctx)
((u64)ctx->irqs.range[3] & 0xffff));
}
-static int attach_dedicated(struct cxl_context *ctx, u64 wed, u64 amr)
+int cxl_attach_dedicated_process_psl(struct cxl_context *ctx, u64 wed, u64 amr)
{
struct cxl_afu *afu = ctx->afu;
u64 pid;
@@ -728,7 +728,8 @@ static int attach_dedicated(struct cxl_context *ctx, u64 wed, u64 amr)
cxl_prefault(ctx, wed);
- update_ivtes_dedicated(ctx);
+ if (ctx->afu->adapter->native->sl_ops->update_dedicated_ivtes)
+ afu->adapter->native->sl_ops->update_dedicated_ivtes(ctx);
cxl_p2n_write(afu, CXL_PSL_AMR_An, amr);
@@ -778,8 +779,9 @@ static int native_afu_activate_mode(struct cxl_afu *afu, int mode)
if (mode == CXL_MODE_DIRECTED)
return activate_afu_directed(afu);
- if (mode == CXL_MODE_DEDICATED)
- return activate_dedicated_process(afu);
+ if ((mode == CXL_MODE_DEDICATED) &&
+ (afu->adapter->native->sl_ops->activate_dedicated_process))
+ return afu->adapter->native->sl_ops->activate_dedicated_process(afu);
return -EINVAL;
}
@@ -793,11 +795,13 @@ static int native_attach_process(struct cxl_context *ctx, bool kernel,
}
ctx->kernel = kernel;
- if (ctx->afu->current_mode == CXL_MODE_DIRECTED)
- return attach_afu_directed(ctx, wed, amr);
+ if ((ctx->afu->current_mode == CXL_MODE_DIRECTED) &&
+ (ctx->afu->adapter->native->sl_ops->attach_afu_directed))
+ return ctx->afu->adapter->native->sl_ops->attach_afu_directed(ctx, wed, amr);
- if (ctx->afu->current_mode == CXL_MODE_DEDICATED)
- return attach_dedicated(ctx, wed, amr);
+ if ((ctx->afu->current_mode == CXL_MODE_DEDICATED) &&
+ (ctx->afu->adapter->native->sl_ops->attach_dedicated_process))
+ return ctx->afu->adapter->native->sl_ops->attach_dedicated_process(ctx, wed, amr);
return -EINVAL;
}
@@ -830,8 +834,9 @@ static void native_update_ivtes(struct cxl_context *ctx)
{
if (ctx->afu->current_mode == CXL_MODE_DIRECTED)
return update_ivtes_directed(ctx);
- if (ctx->afu->current_mode == CXL_MODE_DEDICATED)
- return update_ivtes_dedicated(ctx);
+ if ((ctx->afu->current_mode == CXL_MODE_DEDICATED) &&
+ (ctx->afu->adapter->native->sl_ops->update_dedicated_ivtes))
+ return ctx->afu->adapter->native->sl_ops->update_dedicated_ivtes(ctx);
WARN(1, "native_update_ivtes: Bad mode\n");
}
@@ -875,7 +880,7 @@ static int native_get_irq_info(struct cxl_afu *afu, struct cxl_irq_info *info)
return 0;
}
-void cxl_native_psl_irq_dump_regs(struct cxl_context *ctx)
+void cxl_native_irq_dump_regs_psl(struct cxl_context *ctx)
{
u64 fir1, fir2, fir_slice, serr, afu_debug;
@@ -911,7 +916,7 @@ static irqreturn_t native_handle_psl_slice_error(struct cxl_context *ctx,
return cxl_ops->ack_irq(ctx, 0, errstat);
}
-static irqreturn_t fail_psl_irq(struct cxl_afu *afu, struct cxl_irq_info *irq_info)
+irqreturn_t cxl_fail_irq_psl(struct cxl_afu *afu, struct cxl_irq_info *irq_info)
{
if (irq_info->dsisr & CXL_PSL_DSISR_TRANS)
cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_AE);
@@ -927,7 +932,7 @@ static irqreturn_t native_irq_multiplexed(int irq, void *data)
struct cxl_context *ctx;
struct cxl_irq_info irq_info;
u64 phreg = cxl_p2n_read(afu, CXL_PSL_PEHandle_An);
- int ph, ret;
+ int ph, ret = IRQ_HANDLED;
/* check if eeh kicked in while the interrupt was in flight */
if (unlikely(phreg == ~0ULL)) {
@@ -940,13 +945,16 @@ static irqreturn_t native_irq_multiplexed(int irq, void *data)
ph = phreg & 0xffff;
if ((ret = native_get_irq_info(afu, &irq_info))) {
WARN(1, "Unable to get CXL IRQ Info: %i\n", ret);
- return fail_psl_irq(afu, &irq_info);
+ if (afu->adapter->native->sl_ops->fail_irq)
+ return afu->adapter->native->sl_ops->fail_irq(afu, &irq_info);
+ return IRQ_HANDLED;
}
rcu_read_lock();
ctx = idr_find(&afu->contexts_idr, ph);
if (ctx) {
- ret = cxl_irq(irq, ctx, &irq_info);
+ if (afu->adapter->native->sl_ops->handle_interrupt)
+ ret = afu->adapter->native->sl_ops->handle_interrupt(irq, ctx, &irq_info);
rcu_read_unlock();
return ret;
}
@@ -956,7 +964,9 @@ static irqreturn_t native_irq_multiplexed(int irq, void *data)
" %016llx\n(Possible AFU HW issue - was a term/remove acked"
" with outstanding transactions?)\n", ph, irq_info.dsisr,
irq_info.dar);
- return fail_psl_irq(afu, &irq_info);
+ if (afu->adapter->native->sl_ops->fail_irq)
+ ret = afu->adapter->native->sl_ops->fail_irq(afu, &irq_info);
+ return ret;
}
static void native_irq_wait(struct cxl_context *ctx)
diff --git a/drivers/misc/cxl/pci.c b/drivers/misc/cxl/pci.c
index e82a207..4d475a5 100644
--- a/drivers/misc/cxl/pci.c
+++ b/drivers/misc/cxl/pci.c
@@ -377,7 +377,7 @@ static int calc_capp_routing(struct pci_dev *dev, u64 *chipid, u64 *capp_unit_id
return 0;
}
-static int init_implementation_adapter_psl_regs(struct cxl *adapter, struct pci_dev *dev)
+static int init_implementation_adapter_regs_psl(struct cxl *adapter, struct pci_dev *dev)
{
u64 psl_dsnctl, psl_fircntl;
u64 chipid;
@@ -409,7 +409,7 @@ static int init_implementation_adapter_psl_regs(struct cxl *adapter, struct pci_
return 0;
}
-static int init_implementation_adapter_xsl_regs(struct cxl *adapter, struct pci_dev *dev)
+static int init_implementation_adapter_regs_xsl(struct cxl *adapter, struct pci_dev *dev)
{
u64 xsl_dsnctl;
u64 chipid;
@@ -513,7 +513,7 @@ static void cxl_setup_psl_timebase(struct cxl *adapter, struct pci_dev *dev)
return;
}
-static int init_implementation_afu_psl_regs(struct cxl_afu *afu)
+static int init_implementation_afu_regs_psl(struct cxl_afu *afu)
{
/* read/write masks for this slice */
cxl_p1n_write(afu, CXL_PSL_APCALLOC_A, 0xFFFFFFFEFEFEFEFEULL);
@@ -996,7 +996,7 @@ static int cxl_afu_descriptor_looks_ok(struct cxl_afu *afu)
return 0;
}
-static int sanitise_afu_regs(struct cxl_afu *afu)
+static int sanitise_afu_regs_psl(struct cxl_afu *afu)
{
u64 reg;
@@ -1102,8 +1102,11 @@ static int pci_configure_afu(struct cxl_afu *afu, struct cxl *adapter, struct pc
if ((rc = pci_map_slice_regs(afu, adapter, dev)))
return rc;
- if ((rc = sanitise_afu_regs(afu)))
- goto err1;
+ if (adapter->native->sl_ops->sanitise_afu_regs) {
+ rc = adapter->native->sl_ops->sanitise_afu_regs(afu);
+ if (rc)
+ goto err1;
+ }
/* We need to reset the AFU before we can read the AFU descriptor */
if ((rc = cxl_ops->afu_reset(afu)))
@@ -1432,9 +1435,15 @@ static void cxl_release_adapter(struct device *dev)
static int sanitise_adapter_regs(struct cxl *adapter)
{
+ int rc = 0;
+
/* Clear PSL tberror bit by writing 1 to it */
cxl_p1_write(adapter, CXL_PSL_ErrIVTE, CXL_PSL_ErrIVTE_tberror);
- return cxl_tlb_slb_invalidate(adapter);
+
+ if (adapter->native->sl_ops->invalidate_all)
+ rc = adapter->native->sl_ops->invalidate_all(adapter);
+
+ return rc;
}
/* This should contain *only* operations that can safely be done in
@@ -1518,15 +1527,23 @@ static void cxl_deconfigure_adapter(struct cxl *adapter)
}
static const struct cxl_service_layer_ops psl_ops = {
- .adapter_regs_init = init_implementation_adapter_psl_regs,
- .afu_regs_init = init_implementation_afu_psl_regs,
+ .adapter_regs_init = init_implementation_adapter_regs_psl,
+ .invalidate_all = cxl_invalidate_all_psl,
+ .afu_regs_init = init_implementation_afu_regs_psl,
+ .sanitise_afu_regs = sanitise_afu_regs_psl,
.register_serr_irq = cxl_native_register_serr_irq,
.release_serr_irq = cxl_native_release_serr_irq,
- .debugfs_add_adapter_sl_regs = cxl_debugfs_add_adapter_psl_regs,
- .debugfs_add_afu_sl_regs = cxl_debugfs_add_afu_psl_regs,
- .psl_irq_dump_registers = cxl_native_psl_irq_dump_regs,
+ .handle_interrupt = cxl_irq_psl,
+ .fail_irq = cxl_fail_irq_psl,
+ .activate_dedicated_process = cxl_activate_dedicated_process_psl,
+ .attach_afu_directed = cxl_attach_afu_directed_psl,
+ .attach_dedicated_process = cxl_attach_dedicated_process_psl,
+ .update_dedicated_ivtes = cxl_update_dedicated_ivtes_psl,
+ .debugfs_add_adapter_regs = cxl_debugfs_add_adapter_regs_psl,
+ .debugfs_add_afu_regs = cxl_debugfs_add_afu_regs_psl,
+ .psl_irq_dump_registers = cxl_native_irq_dump_regs_psl,
.err_irq_dump_registers = cxl_native_err_irq_dump_regs,
- .debugfs_stop_trace = cxl_stop_trace,
+ .debugfs_stop_trace = cxl_stop_trace_psl,
.write_timebase_ctrl = write_timebase_ctrl_psl,
.timebase_read = timebase_read_psl,
.capi_mode = OPAL_PHB_CAPI_MODE_CAPI,
@@ -1534,8 +1551,8 @@ static const struct cxl_service_layer_ops psl_ops = {
};
static const struct cxl_service_layer_ops xsl_ops = {
- .adapter_regs_init = init_implementation_adapter_xsl_regs,
- .debugfs_add_adapter_sl_regs = cxl_debugfs_add_adapter_xsl_regs,
+ .adapter_regs_init = init_implementation_adapter_regs_xsl,
+ .debugfs_add_adapter_regs = cxl_debugfs_add_adapter_regs_xsl,
.write_timebase_ctrl = write_timebase_ctrl_xsl,
.timebase_read = timebase_read_xsl,
.capi_mode = OPAL_PHB_CAPI_MODE_DMA,
--
2.7.4
^ permalink raw reply related [flat|nested] 20+ messages in thread* Re: [PATCH V2 3/6] cxl: Update implementation service layer
2017-03-14 11:08 ` [PATCH V2 3/6] cxl: Update implementation service layer Christophe Lombard
@ 2017-03-20 16:17 ` Frederic Barrat
2017-03-21 2:56 ` Andrew Donnellan
0 siblings, 1 reply; 20+ messages in thread
From: Frederic Barrat @ 2017-03-20 16:17 UTC (permalink / raw)
To: Christophe Lombard, linuxppc-dev, imunsie, andrew.donnellan
Le 14/03/2017 à 12:08, Christophe Lombard a écrit :
> The service layer API (in cxl.h) lists some low-level functions whose
> implementation is different on PSL8, PSL9 and XSL:
> - Init implementation for the adapter and the afu.
> - Invalidate TLB/SLB.
> - Attach process for dedicated/directed models.
> - Handle psl interrupts.
> - Debug registers for the adapter and the afu.
> - Traces.
> Each environment implements its own functions, and the common code uses
> them through function pointers, defined in cxl_service_layer_ops.
>
> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com>
> ---
> drivers/misc/cxl/cxl.h | 34 ++++++++++++++++++++++---------
> drivers/misc/cxl/debugfs.c | 16 +++++++--------
> drivers/misc/cxl/guest.c | 2 +-
> drivers/misc/cxl/irq.c | 2 +-
> drivers/misc/cxl/native.c | 50 +++++++++++++++++++++++++++-------------------
> drivers/misc/cxl/pci.c | 47 +++++++++++++++++++++++++++++--------------
> 6 files changed, 97 insertions(+), 54 deletions(-)
>
> diff --git a/drivers/misc/cxl/cxl.h b/drivers/misc/cxl/cxl.h
> index 4d1b704..3e03a66 100644
> --- a/drivers/misc/cxl/cxl.h
> +++ b/drivers/misc/cxl/cxl.h
> @@ -553,13 +553,23 @@ struct cxl_context {
> struct mm_struct *mm;
> };
>
> +struct cxl_irq_info;
> +
> struct cxl_service_layer_ops {
> int (*adapter_regs_init)(struct cxl *adapter, struct pci_dev *dev);
> + int (*invalidate_all)(struct cxl *adapter);
> int (*afu_regs_init)(struct cxl_afu *afu);
> + int (*sanitise_afu_regs)(struct cxl_afu *afu);
> int (*register_serr_irq)(struct cxl_afu *afu);
> void (*release_serr_irq)(struct cxl_afu *afu);
> - void (*debugfs_add_adapter_sl_regs)(struct cxl *adapter, struct dentry *dir);
> - void (*debugfs_add_afu_sl_regs)(struct cxl_afu *afu, struct dentry *dir);
> + irqreturn_t (*handle_interrupt)(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info);
> + irqreturn_t (*fail_irq)(struct cxl_afu *afu, struct cxl_irq_info *irq_info);
> + int (*activate_dedicated_process)(struct cxl_afu *afu);
> + int (*attach_afu_directed)(struct cxl_context *ctx, u64 wed, u64 amr);
> + int (*attach_dedicated_process)(struct cxl_context *ctx, u64 wed, u64 amr);
> + void (*update_dedicated_ivtes)(struct cxl_context *ctx);
> + void (*debugfs_add_adapter_regs)(struct cxl *adapter, struct dentry *dir);
> + void (*debugfs_add_afu_regs)(struct cxl_afu *afu, struct dentry *dir);
> void (*psl_irq_dump_registers)(struct cxl_context *ctx);
> void (*err_irq_dump_registers)(struct cxl *adapter);
> void (*debugfs_stop_trace)(struct cxl *adapter);
> @@ -805,16 +815,21 @@ void afu_irq_name_free(struct cxl_context *ctx);
>
> #ifdef CONFIG_DEBUG_FS
>
> +int cxl_attach_afu_directed_psl(struct cxl_context *ctx, u64 wed, u64 amr);
> +int cxl_activate_dedicated_process_psl(struct cxl_afu *afu);
> +int cxl_attach_dedicated_process_psl(struct cxl_context *ctx, u64 wed, u64 amr);
> +void cxl_update_dedicated_ivtes_psl(struct cxl_context *ctx);
> +
I don't think those prototypes are meant to be dependent on CONFIG_DEBUG_FS
> int cxl_debugfs_init(void);
> void cxl_debugfs_exit(void);
> int cxl_debugfs_adapter_add(struct cxl *adapter);
> void cxl_debugfs_adapter_remove(struct cxl *adapter);
> int cxl_debugfs_afu_add(struct cxl_afu *afu);
> void cxl_debugfs_afu_remove(struct cxl_afu *afu);
> -void cxl_stop_trace(struct cxl *cxl);
> -void cxl_debugfs_add_adapter_psl_regs(struct cxl *adapter, struct dentry *dir);
> -void cxl_debugfs_add_adapter_xsl_regs(struct cxl *adapter, struct dentry *dir);
> -void cxl_debugfs_add_afu_psl_regs(struct cxl_afu *afu, struct dentry *dir);
> +void cxl_stop_trace_psl(struct cxl *cxl);
> +void cxl_debugfs_add_adapter_regs_psl(struct cxl *adapter, struct dentry *dir);
> +void cxl_debugfs_add_adapter_regs_xsl(struct cxl *adapter, struct dentry *dir);
> +void cxl_debugfs_add_afu_regs_psl(struct cxl_afu *afu, struct dentry *dir);
>
> #else /* CONFIG_DEBUG_FS */
>
> @@ -916,19 +931,20 @@ struct cxl_irq_info {
> };
>
> void cxl_assign_psn_space(struct cxl_context *ctx);
> -irqreturn_t cxl_irq(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info);
> +int cxl_invalidate_all_psl(struct cxl *adapter);
> +irqreturn_t cxl_irq_psl(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info);
> +irqreturn_t cxl_fail_irq_psl(struct cxl_afu *afu, struct cxl_irq_info *irq_info);
> int cxl_register_one_irq(struct cxl *adapter, irq_handler_t handler,
> void *cookie, irq_hw_number_t *dest_hwirq,
> unsigned int *dest_virq, const char *name);
>
> int cxl_check_error(struct cxl_afu *afu);
> int cxl_afu_slbia(struct cxl_afu *afu);
> -int cxl_tlb_slb_invalidate(struct cxl *adapter);
> int cxl_data_cache_flush(struct cxl *adapter);
> int cxl_afu_disable(struct cxl_afu *afu);
> int cxl_psl_purge(struct cxl_afu *afu);
>
> -void cxl_native_psl_irq_dump_regs(struct cxl_context *ctx);
> +void cxl_native_irq_dump_regs_psl(struct cxl_context *ctx);
> void cxl_native_err_irq_dump_regs(struct cxl *adapter);
> int cxl_pci_vphb_add(struct cxl_afu *afu);
> void cxl_pci_vphb_remove(struct cxl_afu *afu);
> diff --git a/drivers/misc/cxl/debugfs.c b/drivers/misc/cxl/debugfs.c
> index 9c06ac8..4848ebf 100644
> --- a/drivers/misc/cxl/debugfs.c
> +++ b/drivers/misc/cxl/debugfs.c
> @@ -15,7 +15,7 @@
>
> static struct dentry *cxl_debugfs;
>
> -void cxl_stop_trace(struct cxl *adapter)
> +void cxl_stop_trace_psl(struct cxl *adapter)
> {
> int slice;
>
> @@ -53,7 +53,7 @@ static struct dentry *debugfs_create_io_x64(const char *name, umode_t mode,
> (void __force *)value, &fops_io_x64);
> }
>
> -void cxl_debugfs_add_adapter_psl_regs(struct cxl *adapter, struct dentry *dir)
> +void cxl_debugfs_add_adapter_regs_psl(struct cxl *adapter, struct dentry *dir)
> {
> debugfs_create_io_x64("fir1", S_IRUSR, dir, _cxl_p1_addr(adapter, CXL_PSL_FIR1));
> debugfs_create_io_x64("fir2", S_IRUSR, dir, _cxl_p1_addr(adapter, CXL_PSL_FIR2));
> @@ -61,7 +61,7 @@ void cxl_debugfs_add_adapter_psl_regs(struct cxl *adapter, struct dentry *dir)
> debugfs_create_io_x64("trace", S_IRUSR | S_IWUSR, dir, _cxl_p1_addr(adapter, CXL_PSL_TRACE));
> }
>
> -void cxl_debugfs_add_adapter_xsl_regs(struct cxl *adapter, struct dentry *dir)
> +void cxl_debugfs_add_adapter_regs_xsl(struct cxl *adapter, struct dentry *dir)
> {
> debugfs_create_io_x64("fec", S_IRUSR, dir, _cxl_p1_addr(adapter, CXL_XSL_FEC));
> }
> @@ -82,8 +82,8 @@ int cxl_debugfs_adapter_add(struct cxl *adapter)
>
> debugfs_create_io_x64("err_ivte", S_IRUSR, dir, _cxl_p1_addr(adapter, CXL_PSL_ErrIVTE));
>
> - if (adapter->native->sl_ops->debugfs_add_adapter_sl_regs)
> - adapter->native->sl_ops->debugfs_add_adapter_sl_regs(adapter, dir);
> + if (adapter->native->sl_ops->debugfs_add_adapter_regs)
> + adapter->native->sl_ops->debugfs_add_adapter_regs(adapter, dir);
> return 0;
> }
>
> @@ -92,7 +92,7 @@ void cxl_debugfs_adapter_remove(struct cxl *adapter)
> debugfs_remove_recursive(adapter->debugfs);
> }
>
> -void cxl_debugfs_add_afu_psl_regs(struct cxl_afu *afu, struct dentry *dir)
> +void cxl_debugfs_add_afu_regs_psl(struct cxl_afu *afu, struct dentry *dir)
> {
> debugfs_create_io_x64("fir", S_IRUSR, dir, _cxl_p1n_addr(afu, CXL_PSL_FIR_SLICE_An));
> debugfs_create_io_x64("serr", S_IRUSR, dir, _cxl_p1n_addr(afu, CXL_PSL_SERR_An));
> @@ -121,8 +121,8 @@ int cxl_debugfs_afu_add(struct cxl_afu *afu)
> debugfs_create_io_x64("sstp1", S_IRUSR, dir, _cxl_p2n_addr(afu, CXL_SSTP1_An));
> debugfs_create_io_x64("err_status", S_IRUSR, dir, _cxl_p2n_addr(afu, CXL_PSL_ErrStat_An));
>
> - if (afu->adapter->native->sl_ops->debugfs_add_afu_sl_regs)
> - afu->adapter->native->sl_ops->debugfs_add_afu_sl_regs(afu, dir);
> + if (afu->adapter->native->sl_ops->debugfs_add_afu_regs)
> + afu->adapter->native->sl_ops->debugfs_add_afu_regs(afu, dir);
>
> return 0;
> }
> diff --git a/drivers/misc/cxl/guest.c b/drivers/misc/cxl/guest.c
> index e04bc4d..f6ba698 100644
> --- a/drivers/misc/cxl/guest.c
> +++ b/drivers/misc/cxl/guest.c
> @@ -169,7 +169,7 @@ static irqreturn_t guest_psl_irq(int irq, void *data)
> return IRQ_HANDLED;
> }
>
> - rc = cxl_irq(irq, ctx, &irq_info);
> + rc = cxl_irq_psl(irq, ctx, &irq_info);
> return rc;
> }
>
> diff --git a/drivers/misc/cxl/irq.c b/drivers/misc/cxl/irq.c
> index 1a402bb..2fa119e 100644
> --- a/drivers/misc/cxl/irq.c
> +++ b/drivers/misc/cxl/irq.c
> @@ -34,7 +34,7 @@ static irqreturn_t schedule_cxl_fault(struct cxl_context *ctx, u64 dsisr, u64 da
> return IRQ_HANDLED;
> }
>
> -irqreturn_t cxl_irq(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info)
> +irqreturn_t cxl_irq_psl(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info)
> {
> u64 dsisr, dar;
>
> diff --git a/drivers/misc/cxl/native.c b/drivers/misc/cxl/native.c
> index 7257e8b..2d1b0c6 100644
> --- a/drivers/misc/cxl/native.c
> +++ b/drivers/misc/cxl/native.c
> @@ -258,7 +258,7 @@ void cxl_release_spa(struct cxl_afu *afu)
> }
> }
>
> -int cxl_tlb_slb_invalidate(struct cxl *adapter)
> +int cxl_invalidate_all_psl(struct cxl *adapter)
> {
> unsigned long timeout = jiffies + (HZ * CXL_TIMEOUT);
>
> @@ -578,7 +578,7 @@ static void update_ivtes_directed(struct cxl_context *ctx)
> WARN_ON(add_process_element(ctx));
> }
>
> -static int attach_afu_directed(struct cxl_context *ctx, u64 wed, u64 amr)
> +int cxl_attach_afu_directed_psl(struct cxl_context *ctx, u64 wed, u64 amr)
> {
> u32 pid;
> int result;
> @@ -671,7 +671,7 @@ static int deactivate_afu_directed(struct cxl_afu *afu)
> return 0;
> }
>
> -static int activate_dedicated_process(struct cxl_afu *afu)
> +int cxl_activate_dedicated_process_psl(struct cxl_afu *afu)
> {
> dev_info(&afu->dev, "Activating dedicated process mode\n");
>
> @@ -694,7 +694,7 @@ static int activate_dedicated_process(struct cxl_afu *afu)
> return cxl_chardev_d_afu_add(afu);
> }
>
> -static void update_ivtes_dedicated(struct cxl_context *ctx)
> +void cxl_update_dedicated_ivtes_psl(struct cxl_context *ctx)
> {
> struct cxl_afu *afu = ctx->afu;
>
> @@ -710,7 +710,7 @@ static void update_ivtes_dedicated(struct cxl_context *ctx)
> ((u64)ctx->irqs.range[3] & 0xffff));
> }
>
> -static int attach_dedicated(struct cxl_context *ctx, u64 wed, u64 amr)
> +int cxl_attach_dedicated_process_psl(struct cxl_context *ctx, u64 wed, u64 amr)
> {
> struct cxl_afu *afu = ctx->afu;
> u64 pid;
> @@ -728,7 +728,8 @@ static int attach_dedicated(struct cxl_context *ctx, u64 wed, u64 amr)
>
> cxl_prefault(ctx, wed);
>
> - update_ivtes_dedicated(ctx);
> + if (ctx->afu->adapter->native->sl_ops->update_dedicated_ivtes)
> + afu->adapter->native->sl_ops->update_dedicated_ivtes(ctx);
>
> cxl_p2n_write(afu, CXL_PSL_AMR_An, amr);
>
> @@ -778,8 +779,9 @@ static int native_afu_activate_mode(struct cxl_afu *afu, int mode)
>
> if (mode == CXL_MODE_DIRECTED)
> return activate_afu_directed(afu);
> - if (mode == CXL_MODE_DEDICATED)
> - return activate_dedicated_process(afu);
> + if ((mode == CXL_MODE_DEDICATED) &&
> + (afu->adapter->native->sl_ops->activate_dedicated_process))
> + return afu->adapter->native->sl_ops->activate_dedicated_process(afu);
>
> return -EINVAL;
> }
> @@ -793,11 +795,13 @@ static int native_attach_process(struct cxl_context *ctx, bool kernel,
> }
>
> ctx->kernel = kernel;
> - if (ctx->afu->current_mode == CXL_MODE_DIRECTED)
> - return attach_afu_directed(ctx, wed, amr);
> + if ((ctx->afu->current_mode == CXL_MODE_DIRECTED) &&
> + (ctx->afu->adapter->native->sl_ops->attach_afu_directed))
> + return ctx->afu->adapter->native->sl_ops->attach_afu_directed(ctx, wed, amr);
>
> - if (ctx->afu->current_mode == CXL_MODE_DEDICATED)
> - return attach_dedicated(ctx, wed, amr);
> + if ((ctx->afu->current_mode == CXL_MODE_DEDICATED) &&
> + (ctx->afu->adapter->native->sl_ops->attach_dedicated_process))
> + return ctx->afu->adapter->native->sl_ops->attach_dedicated_process(ctx, wed, amr);
>
> return -EINVAL;
> }
> @@ -830,8 +834,9 @@ static void native_update_ivtes(struct cxl_context *ctx)
> {
> if (ctx->afu->current_mode == CXL_MODE_DIRECTED)
> return update_ivtes_directed(ctx);
> - if (ctx->afu->current_mode == CXL_MODE_DEDICATED)
> - return update_ivtes_dedicated(ctx);
> + if ((ctx->afu->current_mode == CXL_MODE_DEDICATED) &&
> + (ctx->afu->adapter->native->sl_ops->update_dedicated_ivtes))
> + return ctx->afu->adapter->native->sl_ops->update_dedicated_ivtes(ctx);
> WARN(1, "native_update_ivtes: Bad mode\n");
> }
>
> @@ -875,7 +880,7 @@ static int native_get_irq_info(struct cxl_afu *afu, struct cxl_irq_info *info)
> return 0;
> }
>
> -void cxl_native_psl_irq_dump_regs(struct cxl_context *ctx)
> +void cxl_native_irq_dump_regs_psl(struct cxl_context *ctx)
> {
> u64 fir1, fir2, fir_slice, serr, afu_debug;
>
> @@ -911,7 +916,7 @@ static irqreturn_t native_handle_psl_slice_error(struct cxl_context *ctx,
> return cxl_ops->ack_irq(ctx, 0, errstat);
> }
>
> -static irqreturn_t fail_psl_irq(struct cxl_afu *afu, struct cxl_irq_info *irq_info)
> +irqreturn_t cxl_fail_irq_psl(struct cxl_afu *afu, struct cxl_irq_info *irq_info)
> {
> if (irq_info->dsisr & CXL_PSL_DSISR_TRANS)
> cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_AE);
> @@ -927,7 +932,7 @@ static irqreturn_t native_irq_multiplexed(int irq, void *data)
> struct cxl_context *ctx;
> struct cxl_irq_info irq_info;
> u64 phreg = cxl_p2n_read(afu, CXL_PSL_PEHandle_An);
> - int ph, ret;
> + int ph, ret = IRQ_HANDLED;
>
> /* check if eeh kicked in while the interrupt was in flight */
> if (unlikely(phreg == ~0ULL)) {
> @@ -940,13 +945,16 @@ static irqreturn_t native_irq_multiplexed(int irq, void *data)
> ph = phreg & 0xffff;
> if ((ret = native_get_irq_info(afu, &irq_info))) {
> WARN(1, "Unable to get CXL IRQ Info: %i\n", ret);
> - return fail_psl_irq(afu, &irq_info);
> + if (afu->adapter->native->sl_ops->fail_irq)
> + return afu->adapter->native->sl_ops->fail_irq(afu, &irq_info);
> + return IRQ_HANDLED;
> }
>
> rcu_read_lock();
> ctx = idr_find(&afu->contexts_idr, ph);
> if (ctx) {
> - ret = cxl_irq(irq, ctx, &irq_info);
> + if (afu->adapter->native->sl_ops->handle_interrupt)
> + ret = afu->adapter->native->sl_ops->handle_interrupt(irq, ctx, &irq_info);
> rcu_read_unlock();
> return ret;
If the handle_interrupt callback is not defined, we could return here
with ret = 0 (from previous call to native_get_irq_info()) instead of
IRQ_HANDLED.
Fred
> }
> @@ -956,7 +964,9 @@ static irqreturn_t native_irq_multiplexed(int irq, void *data)
> " %016llx\n(Possible AFU HW issue - was a term/remove acked"
> " with outstanding transactions?)\n", ph, irq_info.dsisr,
> irq_info.dar);
> - return fail_psl_irq(afu, &irq_info);
> + if (afu->adapter->native->sl_ops->fail_irq)
> + ret = afu->adapter->native->sl_ops->fail_irq(afu, &irq_info);
> + return ret;
> }
>
> static void native_irq_wait(struct cxl_context *ctx)
> diff --git a/drivers/misc/cxl/pci.c b/drivers/misc/cxl/pci.c
> index e82a207..4d475a5 100644
> --- a/drivers/misc/cxl/pci.c
> +++ b/drivers/misc/cxl/pci.c
> @@ -377,7 +377,7 @@ static int calc_capp_routing(struct pci_dev *dev, u64 *chipid, u64 *capp_unit_id
> return 0;
> }
>
> -static int init_implementation_adapter_psl_regs(struct cxl *adapter, struct pci_dev *dev)
> +static int init_implementation_adapter_regs_psl(struct cxl *adapter, struct pci_dev *dev)
> {
> u64 psl_dsnctl, psl_fircntl;
> u64 chipid;
> @@ -409,7 +409,7 @@ static int init_implementation_adapter_psl_regs(struct cxl *adapter, struct pci_
> return 0;
> }
>
> -static int init_implementation_adapter_xsl_regs(struct cxl *adapter, struct pci_dev *dev)
> +static int init_implementation_adapter_regs_xsl(struct cxl *adapter, struct pci_dev *dev)
> {
> u64 xsl_dsnctl;
> u64 chipid;
> @@ -513,7 +513,7 @@ static void cxl_setup_psl_timebase(struct cxl *adapter, struct pci_dev *dev)
> return;
> }
>
> -static int init_implementation_afu_psl_regs(struct cxl_afu *afu)
> +static int init_implementation_afu_regs_psl(struct cxl_afu *afu)
> {
> /* read/write masks for this slice */
> cxl_p1n_write(afu, CXL_PSL_APCALLOC_A, 0xFFFFFFFEFEFEFEFEULL);
> @@ -996,7 +996,7 @@ static int cxl_afu_descriptor_looks_ok(struct cxl_afu *afu)
> return 0;
> }
>
> -static int sanitise_afu_regs(struct cxl_afu *afu)
> +static int sanitise_afu_regs_psl(struct cxl_afu *afu)
> {
> u64 reg;
>
> @@ -1102,8 +1102,11 @@ static int pci_configure_afu(struct cxl_afu *afu, struct cxl *adapter, struct pc
> if ((rc = pci_map_slice_regs(afu, adapter, dev)))
> return rc;
>
> - if ((rc = sanitise_afu_regs(afu)))
> - goto err1;
> + if (adapter->native->sl_ops->sanitise_afu_regs) {
> + rc = adapter->native->sl_ops->sanitise_afu_regs(afu);
> + if (rc)
> + goto err1;
> + }
>
> /* We need to reset the AFU before we can read the AFU descriptor */
> if ((rc = cxl_ops->afu_reset(afu)))
> @@ -1432,9 +1435,15 @@ static void cxl_release_adapter(struct device *dev)
>
> static int sanitise_adapter_regs(struct cxl *adapter)
> {
> + int rc = 0;
> +
> /* Clear PSL tberror bit by writing 1 to it */
> cxl_p1_write(adapter, CXL_PSL_ErrIVTE, CXL_PSL_ErrIVTE_tberror);
> - return cxl_tlb_slb_invalidate(adapter);
> +
> + if (adapter->native->sl_ops->invalidate_all)
> + rc = adapter->native->sl_ops->invalidate_all(adapter);
> +
> + return rc;
> }
>
> /* This should contain *only* operations that can safely be done in
> @@ -1518,15 +1527,23 @@ static void cxl_deconfigure_adapter(struct cxl *adapter)
> }
>
> static const struct cxl_service_layer_ops psl_ops = {
> - .adapter_regs_init = init_implementation_adapter_psl_regs,
> - .afu_regs_init = init_implementation_afu_psl_regs,
> + .adapter_regs_init = init_implementation_adapter_regs_psl,
> + .invalidate_all = cxl_invalidate_all_psl,
> + .afu_regs_init = init_implementation_afu_regs_psl,
> + .sanitise_afu_regs = sanitise_afu_regs_psl,
> .register_serr_irq = cxl_native_register_serr_irq,
> .release_serr_irq = cxl_native_release_serr_irq,
> - .debugfs_add_adapter_sl_regs = cxl_debugfs_add_adapter_psl_regs,
> - .debugfs_add_afu_sl_regs = cxl_debugfs_add_afu_psl_regs,
> - .psl_irq_dump_registers = cxl_native_psl_irq_dump_regs,
> + .handle_interrupt = cxl_irq_psl,
> + .fail_irq = cxl_fail_irq_psl,
> + .activate_dedicated_process = cxl_activate_dedicated_process_psl,
> + .attach_afu_directed = cxl_attach_afu_directed_psl,
> + .attach_dedicated_process = cxl_attach_dedicated_process_psl,
> + .update_dedicated_ivtes = cxl_update_dedicated_ivtes_psl,
> + .debugfs_add_adapter_regs = cxl_debugfs_add_adapter_regs_psl,
> + .debugfs_add_afu_regs = cxl_debugfs_add_afu_regs_psl,
> + .psl_irq_dump_registers = cxl_native_irq_dump_regs_psl,
> .err_irq_dump_registers = cxl_native_err_irq_dump_regs,
> - .debugfs_stop_trace = cxl_stop_trace,
> + .debugfs_stop_trace = cxl_stop_trace_psl,
> .write_timebase_ctrl = write_timebase_ctrl_psl,
> .timebase_read = timebase_read_psl,
> .capi_mode = OPAL_PHB_CAPI_MODE_CAPI,
> @@ -1534,8 +1551,8 @@ static const struct cxl_service_layer_ops psl_ops = {
> };
>
> static const struct cxl_service_layer_ops xsl_ops = {
> - .adapter_regs_init = init_implementation_adapter_xsl_regs,
> - .debugfs_add_adapter_sl_regs = cxl_debugfs_add_adapter_xsl_regs,
> + .adapter_regs_init = init_implementation_adapter_regs_xsl,
> + .debugfs_add_adapter_regs = cxl_debugfs_add_adapter_regs_xsl,
> .write_timebase_ctrl = write_timebase_ctrl_xsl,
> .timebase_read = timebase_read_xsl,
> .capi_mode = OPAL_PHB_CAPI_MODE_DMA,
>
^ permalink raw reply [flat|nested] 20+ messages in thread* Re: [PATCH V2 3/6] cxl: Update implementation service layer
2017-03-20 16:17 ` Frederic Barrat
@ 2017-03-21 2:56 ` Andrew Donnellan
0 siblings, 0 replies; 20+ messages in thread
From: Andrew Donnellan @ 2017-03-21 2:56 UTC (permalink / raw)
To: Frederic Barrat, Christophe Lombard, linuxppc-dev, imunsie
On 21/03/17 03:17, Frederic Barrat wrote:
>
>
> Le 14/03/2017 à 12:08, Christophe Lombard a écrit :
>> The service layer API (in cxl.h) lists some low-level functions whose
>> implementation is different on PSL8, PSL9 and XSL:
>> - Init implementation for the adapter and the afu.
>> - Invalidate TLB/SLB.
>> - Attach process for dedicated/directed models.
>> - Handle psl interrupts.
>> - Debug registers for the adapter and the afu.
>> - Traces.
>> Each environment implements its own functions, and the common code uses
>> them through function pointers, defined in cxl_service_layer_ops.
>>
>> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com>
>> ---
>> drivers/misc/cxl/cxl.h | 34 ++++++++++++++++++++++---------
>> drivers/misc/cxl/debugfs.c | 16 +++++++--------
>> drivers/misc/cxl/guest.c | 2 +-
>> drivers/misc/cxl/irq.c | 2 +-
>> drivers/misc/cxl/native.c | 50
>> +++++++++++++++++++++++++++-------------------
>> drivers/misc/cxl/pci.c | 47
>> +++++++++++++++++++++++++++++--------------
>> 6 files changed, 97 insertions(+), 54 deletions(-)
>>
>> diff --git a/drivers/misc/cxl/cxl.h b/drivers/misc/cxl/cxl.h
>> index 4d1b704..3e03a66 100644
>> --- a/drivers/misc/cxl/cxl.h
>> +++ b/drivers/misc/cxl/cxl.h
>> @@ -553,13 +553,23 @@ struct cxl_context {
>> struct mm_struct *mm;
>> };
>>
>> +struct cxl_irq_info;
>> +
>> struct cxl_service_layer_ops {
>> int (*adapter_regs_init)(struct cxl *adapter, struct pci_dev *dev);
>> + int (*invalidate_all)(struct cxl *adapter);
>> int (*afu_regs_init)(struct cxl_afu *afu);
>> + int (*sanitise_afu_regs)(struct cxl_afu *afu);
>> int (*register_serr_irq)(struct cxl_afu *afu);
>> void (*release_serr_irq)(struct cxl_afu *afu);
>> - void (*debugfs_add_adapter_sl_regs)(struct cxl *adapter, struct
>> dentry *dir);
>> - void (*debugfs_add_afu_sl_regs)(struct cxl_afu *afu, struct
>> dentry *dir);
>> + irqreturn_t (*handle_interrupt)(int irq, struct cxl_context *ctx,
>> struct cxl_irq_info *irq_info);
>> + irqreturn_t (*fail_irq)(struct cxl_afu *afu, struct cxl_irq_info
>> *irq_info);
>> + int (*activate_dedicated_process)(struct cxl_afu *afu);
>> + int (*attach_afu_directed)(struct cxl_context *ctx, u64 wed, u64
>> amr);
>> + int (*attach_dedicated_process)(struct cxl_context *ctx, u64 wed,
>> u64 amr);
>> + void (*update_dedicated_ivtes)(struct cxl_context *ctx);
>> + void (*debugfs_add_adapter_regs)(struct cxl *adapter, struct
>> dentry *dir);
>> + void (*debugfs_add_afu_regs)(struct cxl_afu *afu, struct dentry
>> *dir);
>> void (*psl_irq_dump_registers)(struct cxl_context *ctx);
>> void (*err_irq_dump_registers)(struct cxl *adapter);
>> void (*debugfs_stop_trace)(struct cxl *adapter);
>> @@ -805,16 +815,21 @@ void afu_irq_name_free(struct cxl_context *ctx);
>>
>> #ifdef CONFIG_DEBUG_FS
>>
>> +int cxl_attach_afu_directed_psl(struct cxl_context *ctx, u64 wed, u64
>> amr);
>> +int cxl_activate_dedicated_process_psl(struct cxl_afu *afu);
>> +int cxl_attach_dedicated_process_psl(struct cxl_context *ctx, u64
>> wed, u64 amr);
>> +void cxl_update_dedicated_ivtes_psl(struct cxl_context *ctx);
>> +
>
>
> I don't think those prototypes are meant to be dependent on CONFIG_DEBUG_FS
Whoops, looks like I screwed that up when I rebased. :/
--
Andrew Donnellan OzLabs, ADL Canberra
andrew.donnellan@au1.ibm.com IBM Australia Limited
^ permalink raw reply [flat|nested] 20+ messages in thread
* [PATCH V2 4/6] cxl: Rename some psl8 specific functions
2017-03-14 11:08 [PATCH V2 0/6] cxl: Add support for Coherent Accelerator Interface Architecture 2.0 Christophe Lombard
` (2 preceding siblings ...)
2017-03-14 11:08 ` [PATCH V2 3/6] cxl: Update implementation service layer Christophe Lombard
@ 2017-03-14 11:08 ` Christophe Lombard
2017-03-20 16:26 ` Frederic Barrat
2017-03-14 11:08 ` [PATCH V2 5/6] cxl: Isolate few psl8 specific calls Christophe Lombard
` (2 subsequent siblings)
6 siblings, 1 reply; 20+ messages in thread
From: Christophe Lombard @ 2017-03-14 11:08 UTC (permalink / raw)
To: linuxppc-dev, fbarrat, imunsie, andrew.donnellan
Rename a few functions, changing the '_psl' suffix to '_psl8', to make
clear that the implementation is psl8 specific.
Those functions will have an equivalent implementation for the psl9 in
a later patch.
Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com>
---
drivers/misc/cxl/cxl.h | 22 ++++++++++-----------
drivers/misc/cxl/debugfs.c | 6 +++---
drivers/misc/cxl/guest.c | 2 +-
drivers/misc/cxl/irq.c | 2 +-
drivers/misc/cxl/native.c | 14 +++++++-------
drivers/misc/cxl/pci.c | 48 +++++++++++++++++++++++-----------------------
6 files changed, 47 insertions(+), 47 deletions(-)
diff --git a/drivers/misc/cxl/cxl.h b/drivers/misc/cxl/cxl.h
index 3e03a66..1f04238 100644
--- a/drivers/misc/cxl/cxl.h
+++ b/drivers/misc/cxl/cxl.h
@@ -815,10 +815,10 @@ void afu_irq_name_free(struct cxl_context *ctx);
#ifdef CONFIG_DEBUG_FS
-int cxl_attach_afu_directed_psl(struct cxl_context *ctx, u64 wed, u64 amr);
-int cxl_activate_dedicated_process_psl(struct cxl_afu *afu);
-int cxl_attach_dedicated_process_psl(struct cxl_context *ctx, u64 wed, u64 amr);
-void cxl_update_dedicated_ivtes_psl(struct cxl_context *ctx);
+int cxl_attach_afu_directed_psl8(struct cxl_context *ctx, u64 wed, u64 amr);
+int cxl_activate_dedicated_process_psl8(struct cxl_afu *afu);
+int cxl_attach_dedicated_process_psl8(struct cxl_context *ctx, u64 wed, u64 amr);
+void cxl_update_dedicated_ivtes_psl8(struct cxl_context *ctx);
int cxl_debugfs_init(void);
void cxl_debugfs_exit(void);
@@ -826,10 +826,10 @@ int cxl_debugfs_adapter_add(struct cxl *adapter);
void cxl_debugfs_adapter_remove(struct cxl *adapter);
int cxl_debugfs_afu_add(struct cxl_afu *afu);
void cxl_debugfs_afu_remove(struct cxl_afu *afu);
-void cxl_stop_trace_psl(struct cxl *cxl);
-void cxl_debugfs_add_adapter_regs_psl(struct cxl *adapter, struct dentry *dir);
+void cxl_stop_trace_psl8(struct cxl *cxl);
+void cxl_debugfs_add_adapter_regs_psl8(struct cxl *adapter, struct dentry *dir);
void cxl_debugfs_add_adapter_regs_xsl(struct cxl *adapter, struct dentry *dir);
-void cxl_debugfs_add_afu_regs_psl(struct cxl_afu *afu, struct dentry *dir);
+void cxl_debugfs_add_afu_regs_psl8(struct cxl_afu *afu, struct dentry *dir);
#else /* CONFIG_DEBUG_FS */
@@ -931,9 +931,9 @@ struct cxl_irq_info {
};
void cxl_assign_psn_space(struct cxl_context *ctx);
-int cxl_invalidate_all_psl(struct cxl *adapter);
-irqreturn_t cxl_irq_psl(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info);
-irqreturn_t cxl_fail_irq_psl(struct cxl_afu *afu, struct cxl_irq_info *irq_info);
+int cxl_invalidate_all_psl8(struct cxl *adapter);
+irqreturn_t cxl_irq_psl8(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info);
+irqreturn_t cxl_fail_irq_psl8(struct cxl_afu *afu, struct cxl_irq_info *irq_info);
int cxl_register_one_irq(struct cxl *adapter, irq_handler_t handler,
void *cookie, irq_hw_number_t *dest_hwirq,
unsigned int *dest_virq, const char *name);
@@ -944,7 +944,7 @@ int cxl_data_cache_flush(struct cxl *adapter);
int cxl_afu_disable(struct cxl_afu *afu);
int cxl_psl_purge(struct cxl_afu *afu);
-void cxl_native_irq_dump_regs_psl(struct cxl_context *ctx);
+void cxl_native_irq_dump_regs_psl8(struct cxl_context *ctx);
void cxl_native_err_irq_dump_regs(struct cxl *adapter);
int cxl_pci_vphb_add(struct cxl_afu *afu);
void cxl_pci_vphb_remove(struct cxl_afu *afu);
diff --git a/drivers/misc/cxl/debugfs.c b/drivers/misc/cxl/debugfs.c
index 4848ebf..2ff10a9 100644
--- a/drivers/misc/cxl/debugfs.c
+++ b/drivers/misc/cxl/debugfs.c
@@ -15,7 +15,7 @@
static struct dentry *cxl_debugfs;
-void cxl_stop_trace_psl(struct cxl *adapter)
+void cxl_stop_trace_psl8(struct cxl *adapter)
{
int slice;
@@ -53,7 +53,7 @@ static struct dentry *debugfs_create_io_x64(const char *name, umode_t mode,
(void __force *)value, &fops_io_x64);
}
-void cxl_debugfs_add_adapter_regs_psl(struct cxl *adapter, struct dentry *dir)
+void cxl_debugfs_add_adapter_regs_psl8(struct cxl *adapter, struct dentry *dir)
{
debugfs_create_io_x64("fir1", S_IRUSR, dir, _cxl_p1_addr(adapter, CXL_PSL_FIR1));
debugfs_create_io_x64("fir2", S_IRUSR, dir, _cxl_p1_addr(adapter, CXL_PSL_FIR2));
@@ -92,7 +92,7 @@ void cxl_debugfs_adapter_remove(struct cxl *adapter)
debugfs_remove_recursive(adapter->debugfs);
}
-void cxl_debugfs_add_afu_regs_psl(struct cxl_afu *afu, struct dentry *dir)
+void cxl_debugfs_add_afu_regs_psl8(struct cxl_afu *afu, struct dentry *dir)
{
debugfs_create_io_x64("fir", S_IRUSR, dir, _cxl_p1n_addr(afu, CXL_PSL_FIR_SLICE_An));
debugfs_create_io_x64("serr", S_IRUSR, dir, _cxl_p1n_addr(afu, CXL_PSL_SERR_An));
diff --git a/drivers/misc/cxl/guest.c b/drivers/misc/cxl/guest.c
index f6ba698..3ad7381 100644
--- a/drivers/misc/cxl/guest.c
+++ b/drivers/misc/cxl/guest.c
@@ -169,7 +169,7 @@ static irqreturn_t guest_psl_irq(int irq, void *data)
return IRQ_HANDLED;
}
- rc = cxl_irq_psl(irq, ctx, &irq_info);
+ rc = cxl_irq_psl8(irq, ctx, &irq_info);
return rc;
}
diff --git a/drivers/misc/cxl/irq.c b/drivers/misc/cxl/irq.c
index 2fa119e..fa9f8a2 100644
--- a/drivers/misc/cxl/irq.c
+++ b/drivers/misc/cxl/irq.c
@@ -34,7 +34,7 @@ static irqreturn_t schedule_cxl_fault(struct cxl_context *ctx, u64 dsisr, u64 da
return IRQ_HANDLED;
}
-irqreturn_t cxl_irq_psl(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info)
+irqreturn_t cxl_irq_psl8(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info)
{
u64 dsisr, dar;
diff --git a/drivers/misc/cxl/native.c b/drivers/misc/cxl/native.c
index 2d1b0c6..ea08605 100644
--- a/drivers/misc/cxl/native.c
+++ b/drivers/misc/cxl/native.c
@@ -258,7 +258,7 @@ void cxl_release_spa(struct cxl_afu *afu)
}
}
-int cxl_invalidate_all_psl(struct cxl *adapter)
+int cxl_invalidate_all_psl8(struct cxl *adapter)
{
unsigned long timeout = jiffies + (HZ * CXL_TIMEOUT);
@@ -578,7 +578,7 @@ static void update_ivtes_directed(struct cxl_context *ctx)
WARN_ON(add_process_element(ctx));
}
-int cxl_attach_afu_directed_psl(struct cxl_context *ctx, u64 wed, u64 amr)
+int cxl_attach_afu_directed_psl8(struct cxl_context *ctx, u64 wed, u64 amr)
{
u32 pid;
int result;
@@ -671,7 +671,7 @@ static int deactivate_afu_directed(struct cxl_afu *afu)
return 0;
}
-int cxl_activate_dedicated_process_psl(struct cxl_afu *afu)
+int cxl_activate_dedicated_process_psl8(struct cxl_afu *afu)
{
dev_info(&afu->dev, "Activating dedicated process mode\n");
@@ -694,7 +694,7 @@ int cxl_activate_dedicated_process_psl(struct cxl_afu *afu)
return cxl_chardev_d_afu_add(afu);
}
-void cxl_update_dedicated_ivtes_psl(struct cxl_context *ctx)
+void cxl_update_dedicated_ivtes_psl8(struct cxl_context *ctx)
{
struct cxl_afu *afu = ctx->afu;
@@ -710,7 +710,7 @@ void cxl_update_dedicated_ivtes_psl(struct cxl_context *ctx)
((u64)ctx->irqs.range[3] & 0xffff));
}
-int cxl_attach_dedicated_process_psl(struct cxl_context *ctx, u64 wed, u64 amr)
+int cxl_attach_dedicated_process_psl8(struct cxl_context *ctx, u64 wed, u64 amr)
{
struct cxl_afu *afu = ctx->afu;
u64 pid;
@@ -880,7 +880,7 @@ static int native_get_irq_info(struct cxl_afu *afu, struct cxl_irq_info *info)
return 0;
}
-void cxl_native_irq_dump_regs_psl(struct cxl_context *ctx)
+void cxl_native_irq_dump_regs_psl8(struct cxl_context *ctx)
{
u64 fir1, fir2, fir_slice, serr, afu_debug;
@@ -916,7 +916,7 @@ static irqreturn_t native_handle_psl_slice_error(struct cxl_context *ctx,
return cxl_ops->ack_irq(ctx, 0, errstat);
}
-irqreturn_t cxl_fail_irq_psl(struct cxl_afu *afu, struct cxl_irq_info *irq_info)
+irqreturn_t cxl_fail_irq_psl8(struct cxl_afu *afu, struct cxl_irq_info *irq_info)
{
if (irq_info->dsisr & CXL_PSL_DSISR_TRANS)
cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_AE);
diff --git a/drivers/misc/cxl/pci.c b/drivers/misc/cxl/pci.c
index 4d475a5..b952527 100644
--- a/drivers/misc/cxl/pci.c
+++ b/drivers/misc/cxl/pci.c
@@ -377,7 +377,7 @@ static int calc_capp_routing(struct pci_dev *dev, u64 *chipid, u64 *capp_unit_id
return 0;
}
-static int init_implementation_adapter_regs_psl(struct cxl *adapter, struct pci_dev *dev)
+static int init_implementation_adapter_regs_psl8(struct cxl *adapter, struct pci_dev *dev)
{
u64 psl_dsnctl, psl_fircntl;
u64 chipid;
@@ -434,7 +434,7 @@ static int init_implementation_adapter_regs_xsl(struct cxl *adapter, struct pci_
/* For the PSL this is a multiple for 0 < n <= 7: */
#define PSL_2048_250MHZ_CYCLES 1
-static void write_timebase_ctrl_psl(struct cxl *adapter)
+static void write_timebase_ctrl_psl8(struct cxl *adapter)
{
cxl_p1_write(adapter, CXL_PSL_TB_CTLSTAT,
TBSYNC_CNT(2 * PSL_2048_250MHZ_CYCLES));
@@ -455,7 +455,7 @@ static void write_timebase_ctrl_xsl(struct cxl *adapter)
TBSYNC_CNT(XSL_4000_CLOCKS));
}
-static u64 timebase_read_psl(struct cxl *adapter)
+static u64 timebase_read_psl8(struct cxl *adapter)
{
return cxl_p1_read(adapter, CXL_PSL_Timebase);
}
@@ -513,7 +513,7 @@ static void cxl_setup_psl_timebase(struct cxl *adapter, struct pci_dev *dev)
return;
}
-static int init_implementation_afu_regs_psl(struct cxl_afu *afu)
+static int init_implementation_afu_regs_psl8(struct cxl_afu *afu)
{
/* read/write masks for this slice */
cxl_p1n_write(afu, CXL_PSL_APCALLOC_A, 0xFFFFFFFEFEFEFEFEULL);
@@ -996,7 +996,7 @@ static int cxl_afu_descriptor_looks_ok(struct cxl_afu *afu)
return 0;
}
-static int sanitise_afu_regs_psl(struct cxl_afu *afu)
+static int sanitise_afu_regs_psl8(struct cxl_afu *afu)
{
u64 reg;
@@ -1526,26 +1526,26 @@ static void cxl_deconfigure_adapter(struct cxl *adapter)
pci_disable_device(pdev);
}
-static const struct cxl_service_layer_ops psl_ops = {
- .adapter_regs_init = init_implementation_adapter_regs_psl,
- .invalidate_all = cxl_invalidate_all_psl,
- .afu_regs_init = init_implementation_afu_regs_psl,
- .sanitise_afu_regs = sanitise_afu_regs_psl,
+static const struct cxl_service_layer_ops psl8_ops = {
+ .adapter_regs_init = init_implementation_adapter_regs_psl8,
+ .invalidate_all = cxl_invalidate_all_psl8,
+ .afu_regs_init = init_implementation_afu_regs_psl8,
+ .sanitise_afu_regs = sanitise_afu_regs_psl8,
.register_serr_irq = cxl_native_register_serr_irq,
.release_serr_irq = cxl_native_release_serr_irq,
- .handle_interrupt = cxl_irq_psl,
- .fail_irq = cxl_fail_irq_psl,
- .activate_dedicated_process = cxl_activate_dedicated_process_psl,
- .attach_afu_directed = cxl_attach_afu_directed_psl,
- .attach_dedicated_process = cxl_attach_dedicated_process_psl,
- .update_dedicated_ivtes = cxl_update_dedicated_ivtes_psl,
- .debugfs_add_adapter_regs = cxl_debugfs_add_adapter_regs_psl,
- .debugfs_add_afu_regs = cxl_debugfs_add_afu_regs_psl,
- .psl_irq_dump_registers = cxl_native_irq_dump_regs_psl,
+ .handle_interrupt = cxl_irq_psl8,
+ .fail_irq = cxl_fail_irq_psl8,
+ .activate_dedicated_process = cxl_activate_dedicated_process_psl8,
+ .attach_afu_directed = cxl_attach_afu_directed_psl8,
+ .attach_dedicated_process = cxl_attach_dedicated_process_psl8,
+ .update_dedicated_ivtes = cxl_update_dedicated_ivtes_psl8,
+ .debugfs_add_adapter_regs = cxl_debugfs_add_adapter_regs_psl8,
+ .debugfs_add_afu_regs = cxl_debugfs_add_afu_regs_psl8,
+ .psl_irq_dump_registers = cxl_native_irq_dump_regs_psl8,
.err_irq_dump_registers = cxl_native_err_irq_dump_regs,
- .debugfs_stop_trace = cxl_stop_trace_psl,
- .write_timebase_ctrl = write_timebase_ctrl_psl,
- .timebase_read = timebase_read_psl,
+ .debugfs_stop_trace = cxl_stop_trace_psl8,
+ .write_timebase_ctrl = write_timebase_ctrl_psl8,
+ .timebase_read = timebase_read_psl8,
.capi_mode = OPAL_PHB_CAPI_MODE_CAPI,
.needs_reset_before_disable = true,
};
@@ -1566,8 +1566,8 @@ static void set_sl_ops(struct cxl *adapter, struct pci_dev *dev)
adapter->native->sl_ops = &xsl_ops;
adapter->min_pe = 1; /* Workaround for CX-4 hardware bug */
} else {
- dev_info(&dev->dev, "Device uses a PSL\n");
- adapter->native->sl_ops = &psl_ops;
+ dev_info(&dev->dev, "Device uses a PSL8\n");
+ adapter->native->sl_ops = &psl8_ops;
}
}
--
2.7.4
^ permalink raw reply related [flat|nested] 20+ messages in thread* Re: [PATCH V2 4/6] cxl: Rename some psl8 specific functions
2017-03-14 11:08 ` [PATCH V2 4/6] cxl: Rename some psl8 specific functions Christophe Lombard
@ 2017-03-20 16:26 ` Frederic Barrat
2017-03-21 9:38 ` christophe lombard
0 siblings, 1 reply; 20+ messages in thread
From: Frederic Barrat @ 2017-03-20 16:26 UTC (permalink / raw)
To: Christophe Lombard, linuxppc-dev, imunsie, andrew.donnellan
Le 14/03/2017 à 12:08, Christophe Lombard a écrit :
> Rename a few functions, changing the '_psl' suffix to '_psl8', to make
> clear that the implementation is psl8 specific.
> Those functions will have an equivalent implementation for the psl9 in
> a later patch.
>
> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com>
> ---
> drivers/misc/cxl/cxl.h | 22 ++++++++++-----------
> drivers/misc/cxl/debugfs.c | 6 +++---
> drivers/misc/cxl/guest.c | 2 +-
> drivers/misc/cxl/irq.c | 2 +-
> drivers/misc/cxl/native.c | 14 +++++++-------
> drivers/misc/cxl/pci.c | 48 +++++++++++++++++++++++-----------------------
> 6 files changed, 47 insertions(+), 47 deletions(-)
>
> diff --git a/drivers/misc/cxl/cxl.h b/drivers/misc/cxl/cxl.h
> index 3e03a66..1f04238 100644
> --- a/drivers/misc/cxl/cxl.h
> +++ b/drivers/misc/cxl/cxl.h
> @@ -815,10 +815,10 @@ void afu_irq_name_free(struct cxl_context *ctx);
>
> #ifdef CONFIG_DEBUG_FS
>
> -int cxl_attach_afu_directed_psl(struct cxl_context *ctx, u64 wed, u64 amr);
> -int cxl_activate_dedicated_process_psl(struct cxl_afu *afu);
> -int cxl_attach_dedicated_process_psl(struct cxl_context *ctx, u64 wed, u64 amr);
> -void cxl_update_dedicated_ivtes_psl(struct cxl_context *ctx);
> +int cxl_attach_afu_directed_psl8(struct cxl_context *ctx, u64 wed, u64 amr);
> +int cxl_activate_dedicated_process_psl8(struct cxl_afu *afu);
> +int cxl_attach_dedicated_process_psl8(struct cxl_context *ctx, u64 wed, u64 amr);
> +void cxl_update_dedicated_ivtes_psl8(struct cxl_context *ctx);
See previous comment regarding "#ifdef CONFIG_DEBUG_FS"
The rest of the patch looks good to me.
Fred
>
> int cxl_debugfs_init(void);
> void cxl_debugfs_exit(void);
> @@ -826,10 +826,10 @@ int cxl_debugfs_adapter_add(struct cxl *adapter);
> void cxl_debugfs_adapter_remove(struct cxl *adapter);
> int cxl_debugfs_afu_add(struct cxl_afu *afu);
> void cxl_debugfs_afu_remove(struct cxl_afu *afu);
> -void cxl_stop_trace_psl(struct cxl *cxl);
> -void cxl_debugfs_add_adapter_regs_psl(struct cxl *adapter, struct dentry *dir);
> +void cxl_stop_trace_psl8(struct cxl *cxl);
> +void cxl_debugfs_add_adapter_regs_psl8(struct cxl *adapter, struct dentry *dir);
> void cxl_debugfs_add_adapter_regs_xsl(struct cxl *adapter, struct dentry *dir);
> -void cxl_debugfs_add_afu_regs_psl(struct cxl_afu *afu, struct dentry *dir);
> +void cxl_debugfs_add_afu_regs_psl8(struct cxl_afu *afu, struct dentry *dir);
>
> #else /* CONFIG_DEBUG_FS */
>
> @@ -931,9 +931,9 @@ struct cxl_irq_info {
> };
>
> void cxl_assign_psn_space(struct cxl_context *ctx);
> -int cxl_invalidate_all_psl(struct cxl *adapter);
> -irqreturn_t cxl_irq_psl(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info);
> -irqreturn_t cxl_fail_irq_psl(struct cxl_afu *afu, struct cxl_irq_info *irq_info);
> +int cxl_invalidate_all_psl8(struct cxl *adapter);
> +irqreturn_t cxl_irq_psl8(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info);
> +irqreturn_t cxl_fail_irq_psl8(struct cxl_afu *afu, struct cxl_irq_info *irq_info);
> int cxl_register_one_irq(struct cxl *adapter, irq_handler_t handler,
> void *cookie, irq_hw_number_t *dest_hwirq,
> unsigned int *dest_virq, const char *name);
> @@ -944,7 +944,7 @@ int cxl_data_cache_flush(struct cxl *adapter);
> int cxl_afu_disable(struct cxl_afu *afu);
> int cxl_psl_purge(struct cxl_afu *afu);
>
> -void cxl_native_irq_dump_regs_psl(struct cxl_context *ctx);
> +void cxl_native_irq_dump_regs_psl8(struct cxl_context *ctx);
> void cxl_native_err_irq_dump_regs(struct cxl *adapter);
> int cxl_pci_vphb_add(struct cxl_afu *afu);
> void cxl_pci_vphb_remove(struct cxl_afu *afu);
> diff --git a/drivers/misc/cxl/debugfs.c b/drivers/misc/cxl/debugfs.c
> index 4848ebf..2ff10a9 100644
> --- a/drivers/misc/cxl/debugfs.c
> +++ b/drivers/misc/cxl/debugfs.c
> @@ -15,7 +15,7 @@
>
> static struct dentry *cxl_debugfs;
>
> -void cxl_stop_trace_psl(struct cxl *adapter)
> +void cxl_stop_trace_psl8(struct cxl *adapter)
> {
> int slice;
>
> @@ -53,7 +53,7 @@ static struct dentry *debugfs_create_io_x64(const char *name, umode_t mode,
> (void __force *)value, &fops_io_x64);
> }
>
> -void cxl_debugfs_add_adapter_regs_psl(struct cxl *adapter, struct dentry *dir)
> +void cxl_debugfs_add_adapter_regs_psl8(struct cxl *adapter, struct dentry *dir)
> {
> debugfs_create_io_x64("fir1", S_IRUSR, dir, _cxl_p1_addr(adapter, CXL_PSL_FIR1));
> debugfs_create_io_x64("fir2", S_IRUSR, dir, _cxl_p1_addr(adapter, CXL_PSL_FIR2));
> @@ -92,7 +92,7 @@ void cxl_debugfs_adapter_remove(struct cxl *adapter)
> debugfs_remove_recursive(adapter->debugfs);
> }
>
> -void cxl_debugfs_add_afu_regs_psl(struct cxl_afu *afu, struct dentry *dir)
> +void cxl_debugfs_add_afu_regs_psl8(struct cxl_afu *afu, struct dentry *dir)
> {
> debugfs_create_io_x64("fir", S_IRUSR, dir, _cxl_p1n_addr(afu, CXL_PSL_FIR_SLICE_An));
> debugfs_create_io_x64("serr", S_IRUSR, dir, _cxl_p1n_addr(afu, CXL_PSL_SERR_An));
> diff --git a/drivers/misc/cxl/guest.c b/drivers/misc/cxl/guest.c
> index f6ba698..3ad7381 100644
> --- a/drivers/misc/cxl/guest.c
> +++ b/drivers/misc/cxl/guest.c
> @@ -169,7 +169,7 @@ static irqreturn_t guest_psl_irq(int irq, void *data)
> return IRQ_HANDLED;
> }
>
> - rc = cxl_irq_psl(irq, ctx, &irq_info);
> + rc = cxl_irq_psl8(irq, ctx, &irq_info);
> return rc;
> }
>
> diff --git a/drivers/misc/cxl/irq.c b/drivers/misc/cxl/irq.c
> index 2fa119e..fa9f8a2 100644
> --- a/drivers/misc/cxl/irq.c
> +++ b/drivers/misc/cxl/irq.c
> @@ -34,7 +34,7 @@ static irqreturn_t schedule_cxl_fault(struct cxl_context *ctx, u64 dsisr, u64 da
> return IRQ_HANDLED;
> }
>
> -irqreturn_t cxl_irq_psl(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info)
> +irqreturn_t cxl_irq_psl8(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info)
> {
> u64 dsisr, dar;
>
> diff --git a/drivers/misc/cxl/native.c b/drivers/misc/cxl/native.c
> index 2d1b0c6..ea08605 100644
> --- a/drivers/misc/cxl/native.c
> +++ b/drivers/misc/cxl/native.c
> @@ -258,7 +258,7 @@ void cxl_release_spa(struct cxl_afu *afu)
> }
> }
>
> -int cxl_invalidate_all_psl(struct cxl *adapter)
> +int cxl_invalidate_all_psl8(struct cxl *adapter)
> {
> unsigned long timeout = jiffies + (HZ * CXL_TIMEOUT);
>
> @@ -578,7 +578,7 @@ static void update_ivtes_directed(struct cxl_context *ctx)
> WARN_ON(add_process_element(ctx));
> }
>
> -int cxl_attach_afu_directed_psl(struct cxl_context *ctx, u64 wed, u64 amr)
> +int cxl_attach_afu_directed_psl8(struct cxl_context *ctx, u64 wed, u64 amr)
> {
> u32 pid;
> int result;
> @@ -671,7 +671,7 @@ static int deactivate_afu_directed(struct cxl_afu *afu)
> return 0;
> }
>
> -int cxl_activate_dedicated_process_psl(struct cxl_afu *afu)
> +int cxl_activate_dedicated_process_psl8(struct cxl_afu *afu)
> {
> dev_info(&afu->dev, "Activating dedicated process mode\n");
>
> @@ -694,7 +694,7 @@ int cxl_activate_dedicated_process_psl(struct cxl_afu *afu)
> return cxl_chardev_d_afu_add(afu);
> }
>
> -void cxl_update_dedicated_ivtes_psl(struct cxl_context *ctx)
> +void cxl_update_dedicated_ivtes_psl8(struct cxl_context *ctx)
> {
> struct cxl_afu *afu = ctx->afu;
>
> @@ -710,7 +710,7 @@ void cxl_update_dedicated_ivtes_psl(struct cxl_context *ctx)
> ((u64)ctx->irqs.range[3] & 0xffff));
> }
>
> -int cxl_attach_dedicated_process_psl(struct cxl_context *ctx, u64 wed, u64 amr)
> +int cxl_attach_dedicated_process_psl8(struct cxl_context *ctx, u64 wed, u64 amr)
> {
> struct cxl_afu *afu = ctx->afu;
> u64 pid;
> @@ -880,7 +880,7 @@ static int native_get_irq_info(struct cxl_afu *afu, struct cxl_irq_info *info)
> return 0;
> }
>
> -void cxl_native_irq_dump_regs_psl(struct cxl_context *ctx)
> +void cxl_native_irq_dump_regs_psl8(struct cxl_context *ctx)
> {
> u64 fir1, fir2, fir_slice, serr, afu_debug;
>
> @@ -916,7 +916,7 @@ static irqreturn_t native_handle_psl_slice_error(struct cxl_context *ctx,
> return cxl_ops->ack_irq(ctx, 0, errstat);
> }
>
> -irqreturn_t cxl_fail_irq_psl(struct cxl_afu *afu, struct cxl_irq_info *irq_info)
> +irqreturn_t cxl_fail_irq_psl8(struct cxl_afu *afu, struct cxl_irq_info *irq_info)
> {
> if (irq_info->dsisr & CXL_PSL_DSISR_TRANS)
> cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_AE);
> diff --git a/drivers/misc/cxl/pci.c b/drivers/misc/cxl/pci.c
> index 4d475a5..b952527 100644
> --- a/drivers/misc/cxl/pci.c
> +++ b/drivers/misc/cxl/pci.c
> @@ -377,7 +377,7 @@ static int calc_capp_routing(struct pci_dev *dev, u64 *chipid, u64 *capp_unit_id
> return 0;
> }
>
> -static int init_implementation_adapter_regs_psl(struct cxl *adapter, struct pci_dev *dev)
> +static int init_implementation_adapter_regs_psl8(struct cxl *adapter, struct pci_dev *dev)
> {
> u64 psl_dsnctl, psl_fircntl;
> u64 chipid;
> @@ -434,7 +434,7 @@ static int init_implementation_adapter_regs_xsl(struct cxl *adapter, struct pci_
> /* For the PSL this is a multiple for 0 < n <= 7: */
> #define PSL_2048_250MHZ_CYCLES 1
>
> -static void write_timebase_ctrl_psl(struct cxl *adapter)
> +static void write_timebase_ctrl_psl8(struct cxl *adapter)
> {
> cxl_p1_write(adapter, CXL_PSL_TB_CTLSTAT,
> TBSYNC_CNT(2 * PSL_2048_250MHZ_CYCLES));
> @@ -455,7 +455,7 @@ static void write_timebase_ctrl_xsl(struct cxl *adapter)
> TBSYNC_CNT(XSL_4000_CLOCKS));
> }
>
> -static u64 timebase_read_psl(struct cxl *adapter)
> +static u64 timebase_read_psl8(struct cxl *adapter)
> {
> return cxl_p1_read(adapter, CXL_PSL_Timebase);
> }
> @@ -513,7 +513,7 @@ static void cxl_setup_psl_timebase(struct cxl *adapter, struct pci_dev *dev)
> return;
> }
>
> -static int init_implementation_afu_regs_psl(struct cxl_afu *afu)
> +static int init_implementation_afu_regs_psl8(struct cxl_afu *afu)
> {
> /* read/write masks for this slice */
> cxl_p1n_write(afu, CXL_PSL_APCALLOC_A, 0xFFFFFFFEFEFEFEFEULL);
> @@ -996,7 +996,7 @@ static int cxl_afu_descriptor_looks_ok(struct cxl_afu *afu)
> return 0;
> }
>
> -static int sanitise_afu_regs_psl(struct cxl_afu *afu)
> +static int sanitise_afu_regs_psl8(struct cxl_afu *afu)
> {
> u64 reg;
>
> @@ -1526,26 +1526,26 @@ static void cxl_deconfigure_adapter(struct cxl *adapter)
> pci_disable_device(pdev);
> }
>
> -static const struct cxl_service_layer_ops psl_ops = {
> - .adapter_regs_init = init_implementation_adapter_regs_psl,
> - .invalidate_all = cxl_invalidate_all_psl,
> - .afu_regs_init = init_implementation_afu_regs_psl,
> - .sanitise_afu_regs = sanitise_afu_regs_psl,
> +static const struct cxl_service_layer_ops psl8_ops = {
> + .adapter_regs_init = init_implementation_adapter_regs_psl8,
> + .invalidate_all = cxl_invalidate_all_psl8,
> + .afu_regs_init = init_implementation_afu_regs_psl8,
> + .sanitise_afu_regs = sanitise_afu_regs_psl8,
> .register_serr_irq = cxl_native_register_serr_irq,
> .release_serr_irq = cxl_native_release_serr_irq,
> - .handle_interrupt = cxl_irq_psl,
> - .fail_irq = cxl_fail_irq_psl,
> - .activate_dedicated_process = cxl_activate_dedicated_process_psl,
> - .attach_afu_directed = cxl_attach_afu_directed_psl,
> - .attach_dedicated_process = cxl_attach_dedicated_process_psl,
> - .update_dedicated_ivtes = cxl_update_dedicated_ivtes_psl,
> - .debugfs_add_adapter_regs = cxl_debugfs_add_adapter_regs_psl,
> - .debugfs_add_afu_regs = cxl_debugfs_add_afu_regs_psl,
> - .psl_irq_dump_registers = cxl_native_irq_dump_regs_psl,
> + .handle_interrupt = cxl_irq_psl8,
> + .fail_irq = cxl_fail_irq_psl8,
> + .activate_dedicated_process = cxl_activate_dedicated_process_psl8,
> + .attach_afu_directed = cxl_attach_afu_directed_psl8,
> + .attach_dedicated_process = cxl_attach_dedicated_process_psl8,
> + .update_dedicated_ivtes = cxl_update_dedicated_ivtes_psl8,
> + .debugfs_add_adapter_regs = cxl_debugfs_add_adapter_regs_psl8,
> + .debugfs_add_afu_regs = cxl_debugfs_add_afu_regs_psl8,
> + .psl_irq_dump_registers = cxl_native_irq_dump_regs_psl8,
> .err_irq_dump_registers = cxl_native_err_irq_dump_regs,
> - .debugfs_stop_trace = cxl_stop_trace_psl,
> - .write_timebase_ctrl = write_timebase_ctrl_psl,
> - .timebase_read = timebase_read_psl,
> + .debugfs_stop_trace = cxl_stop_trace_psl8,
> + .write_timebase_ctrl = write_timebase_ctrl_psl8,
> + .timebase_read = timebase_read_psl8,
> .capi_mode = OPAL_PHB_CAPI_MODE_CAPI,
> .needs_reset_before_disable = true,
> };
> @@ -1566,8 +1566,8 @@ static void set_sl_ops(struct cxl *adapter, struct pci_dev *dev)
> adapter->native->sl_ops = &xsl_ops;
> adapter->min_pe = 1; /* Workaround for CX-4 hardware bug */
> } else {
> - dev_info(&dev->dev, "Device uses a PSL\n");
> - adapter->native->sl_ops = &psl_ops;
> + dev_info(&dev->dev, "Device uses a PSL8\n");
> + adapter->native->sl_ops = &psl8_ops;
> }
> }
>
^ permalink raw reply [flat|nested] 20+ messages in thread* Re: [PATCH V2 4/6] cxl: Rename some psl8 specific functions
2017-03-20 16:26 ` Frederic Barrat
@ 2017-03-21 9:38 ` christophe lombard
0 siblings, 0 replies; 20+ messages in thread
From: christophe lombard @ 2017-03-21 9:38 UTC (permalink / raw)
To: Frederic Barrat, linuxppc-dev, andrew.donnellan
Le 20/03/2017 à 17:26, Frederic Barrat a écrit :
>
>
> Le 14/03/2017 à 12:08, Christophe Lombard a écrit :
>> Rename a few functions, changing the '_psl' suffix to '_psl8', to make
>> clear that the implementation is psl8 specific.
>> Those functions will have an equivalent implementation for the psl9 in
>> a later patch.
>>
>> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com>
>> ---
>> drivers/misc/cxl/cxl.h | 22 ++++++++++-----------
>> drivers/misc/cxl/debugfs.c | 6 +++---
>> drivers/misc/cxl/guest.c | 2 +-
>> drivers/misc/cxl/irq.c | 2 +-
>> drivers/misc/cxl/native.c | 14 +++++++-------
>> drivers/misc/cxl/pci.c | 48
>> +++++++++++++++++++++++-----------------------
>> 6 files changed, 47 insertions(+), 47 deletions(-)
>>
>> diff --git a/drivers/misc/cxl/cxl.h b/drivers/misc/cxl/cxl.h
>> index 3e03a66..1f04238 100644
>> --- a/drivers/misc/cxl/cxl.h
>> +++ b/drivers/misc/cxl/cxl.h
>> @@ -815,10 +815,10 @@ void afu_irq_name_free(struct cxl_context *ctx);
>>
>> #ifdef CONFIG_DEBUG_FS
>>
>> -int cxl_attach_afu_directed_psl(struct cxl_context *ctx, u64 wed,
>> u64 amr);
>> -int cxl_activate_dedicated_process_psl(struct cxl_afu *afu);
>> -int cxl_attach_dedicated_process_psl(struct cxl_context *ctx, u64
>> wed, u64 amr);
>> -void cxl_update_dedicated_ivtes_psl(struct cxl_context *ctx);
>> +int cxl_attach_afu_directed_psl8(struct cxl_context *ctx, u64 wed,
>> u64 amr);
>> +int cxl_activate_dedicated_process_psl8(struct cxl_afu *afu);
>> +int cxl_attach_dedicated_process_psl8(struct cxl_context *ctx, u64
>> wed, u64 amr);
>> +void cxl_update_dedicated_ivtes_psl8(struct cxl_context *ctx);
>
>
> See previous comment regarding "#ifdef CONFIG_DEBUG_FS"
>
> The rest of the patch looks good to me.
>
> Fred
>
>
you are right. Thanks
>
>
>
>>
>> int cxl_debugfs_init(void);
>> void cxl_debugfs_exit(void);
>> @@ -826,10 +826,10 @@ int cxl_debugfs_adapter_add(struct cxl *adapter);
>> void cxl_debugfs_adapter_remove(struct cxl *adapter);
>> int cxl_debugfs_afu_add(struct cxl_afu *afu);
>> void cxl_debugfs_afu_remove(struct cxl_afu *afu);
>> -void cxl_stop_trace_psl(struct cxl *cxl);
>> -void cxl_debugfs_add_adapter_regs_psl(struct cxl *adapter, struct
>> dentry *dir);
>> +void cxl_stop_trace_psl8(struct cxl *cxl);
>> +void cxl_debugfs_add_adapter_regs_psl8(struct cxl *adapter, struct
>> dentry *dir);
>> void cxl_debugfs_add_adapter_regs_xsl(struct cxl *adapter, struct
>> dentry *dir);
>> -void cxl_debugfs_add_afu_regs_psl(struct cxl_afu *afu, struct dentry
>> *dir);
>> +void cxl_debugfs_add_afu_regs_psl8(struct cxl_afu *afu, struct
>> dentry *dir);
>>
>> #else /* CONFIG_DEBUG_FS */
>>
>> @@ -931,9 +931,9 @@ struct cxl_irq_info {
>> };
>>
>> void cxl_assign_psn_space(struct cxl_context *ctx);
>> -int cxl_invalidate_all_psl(struct cxl *adapter);
>> -irqreturn_t cxl_irq_psl(int irq, struct cxl_context *ctx, struct
>> cxl_irq_info *irq_info);
>> -irqreturn_t cxl_fail_irq_psl(struct cxl_afu *afu, struct
>> cxl_irq_info *irq_info);
>> +int cxl_invalidate_all_psl8(struct cxl *adapter);
>> +irqreturn_t cxl_irq_psl8(int irq, struct cxl_context *ctx, struct
>> cxl_irq_info *irq_info);
>> +irqreturn_t cxl_fail_irq_psl8(struct cxl_afu *afu, struct
>> cxl_irq_info *irq_info);
>> int cxl_register_one_irq(struct cxl *adapter, irq_handler_t handler,
>> void *cookie, irq_hw_number_t *dest_hwirq,
>> unsigned int *dest_virq, const char *name);
>> @@ -944,7 +944,7 @@ int cxl_data_cache_flush(struct cxl *adapter);
>> int cxl_afu_disable(struct cxl_afu *afu);
>> int cxl_psl_purge(struct cxl_afu *afu);
>>
>> -void cxl_native_irq_dump_regs_psl(struct cxl_context *ctx);
>> +void cxl_native_irq_dump_regs_psl8(struct cxl_context *ctx);
>> void cxl_native_err_irq_dump_regs(struct cxl *adapter);
>> int cxl_pci_vphb_add(struct cxl_afu *afu);
>> void cxl_pci_vphb_remove(struct cxl_afu *afu);
>> diff --git a/drivers/misc/cxl/debugfs.c b/drivers/misc/cxl/debugfs.c
>> index 4848ebf..2ff10a9 100644
>> --- a/drivers/misc/cxl/debugfs.c
>> +++ b/drivers/misc/cxl/debugfs.c
>> @@ -15,7 +15,7 @@
>>
>> static struct dentry *cxl_debugfs;
>>
>> -void cxl_stop_trace_psl(struct cxl *adapter)
>> +void cxl_stop_trace_psl8(struct cxl *adapter)
>> {
>> int slice;
>>
>> @@ -53,7 +53,7 @@ static struct dentry *debugfs_create_io_x64(const
>> char *name, umode_t mode,
>> (void __force *)value, &fops_io_x64);
>> }
>>
>> -void cxl_debugfs_add_adapter_regs_psl(struct cxl *adapter, struct
>> dentry *dir)
>> +void cxl_debugfs_add_adapter_regs_psl8(struct cxl *adapter, struct
>> dentry *dir)
>> {
>> debugfs_create_io_x64("fir1", S_IRUSR, dir,
>> _cxl_p1_addr(adapter, CXL_PSL_FIR1));
>> debugfs_create_io_x64("fir2", S_IRUSR, dir,
>> _cxl_p1_addr(adapter, CXL_PSL_FIR2));
>> @@ -92,7 +92,7 @@ void cxl_debugfs_adapter_remove(struct cxl *adapter)
>> debugfs_remove_recursive(adapter->debugfs);
>> }
>>
>> -void cxl_debugfs_add_afu_regs_psl(struct cxl_afu *afu, struct dentry
>> *dir)
>> +void cxl_debugfs_add_afu_regs_psl8(struct cxl_afu *afu, struct
>> dentry *dir)
>> {
>> debugfs_create_io_x64("fir", S_IRUSR, dir, _cxl_p1n_addr(afu,
>> CXL_PSL_FIR_SLICE_An));
>> debugfs_create_io_x64("serr", S_IRUSR, dir, _cxl_p1n_addr(afu,
>> CXL_PSL_SERR_An));
>> diff --git a/drivers/misc/cxl/guest.c b/drivers/misc/cxl/guest.c
>> index f6ba698..3ad7381 100644
>> --- a/drivers/misc/cxl/guest.c
>> +++ b/drivers/misc/cxl/guest.c
>> @@ -169,7 +169,7 @@ static irqreturn_t guest_psl_irq(int irq, void
>> *data)
>> return IRQ_HANDLED;
>> }
>>
>> - rc = cxl_irq_psl(irq, ctx, &irq_info);
>> + rc = cxl_irq_psl8(irq, ctx, &irq_info);
>> return rc;
>> }
>>
>> diff --git a/drivers/misc/cxl/irq.c b/drivers/misc/cxl/irq.c
>> index 2fa119e..fa9f8a2 100644
>> --- a/drivers/misc/cxl/irq.c
>> +++ b/drivers/misc/cxl/irq.c
>> @@ -34,7 +34,7 @@ static irqreturn_t schedule_cxl_fault(struct
>> cxl_context *ctx, u64 dsisr, u64 da
>> return IRQ_HANDLED;
>> }
>>
>> -irqreturn_t cxl_irq_psl(int irq, struct cxl_context *ctx, struct
>> cxl_irq_info *irq_info)
>> +irqreturn_t cxl_irq_psl8(int irq, struct cxl_context *ctx, struct
>> cxl_irq_info *irq_info)
>> {
>> u64 dsisr, dar;
>>
>> diff --git a/drivers/misc/cxl/native.c b/drivers/misc/cxl/native.c
>> index 2d1b0c6..ea08605 100644
>> --- a/drivers/misc/cxl/native.c
>> +++ b/drivers/misc/cxl/native.c
>> @@ -258,7 +258,7 @@ void cxl_release_spa(struct cxl_afu *afu)
>> }
>> }
>>
>> -int cxl_invalidate_all_psl(struct cxl *adapter)
>> +int cxl_invalidate_all_psl8(struct cxl *adapter)
>> {
>> unsigned long timeout = jiffies + (HZ * CXL_TIMEOUT);
>>
>> @@ -578,7 +578,7 @@ static void update_ivtes_directed(struct
>> cxl_context *ctx)
>> WARN_ON(add_process_element(ctx));
>> }
>>
>> -int cxl_attach_afu_directed_psl(struct cxl_context *ctx, u64 wed,
>> u64 amr)
>> +int cxl_attach_afu_directed_psl8(struct cxl_context *ctx, u64 wed,
>> u64 amr)
>> {
>> u32 pid;
>> int result;
>> @@ -671,7 +671,7 @@ static int deactivate_afu_directed(struct cxl_afu
>> *afu)
>> return 0;
>> }
>>
>> -int cxl_activate_dedicated_process_psl(struct cxl_afu *afu)
>> +int cxl_activate_dedicated_process_psl8(struct cxl_afu *afu)
>> {
>> dev_info(&afu->dev, "Activating dedicated process mode\n");
>>
>> @@ -694,7 +694,7 @@ int cxl_activate_dedicated_process_psl(struct
>> cxl_afu *afu)
>> return cxl_chardev_d_afu_add(afu);
>> }
>>
>> -void cxl_update_dedicated_ivtes_psl(struct cxl_context *ctx)
>> +void cxl_update_dedicated_ivtes_psl8(struct cxl_context *ctx)
>> {
>> struct cxl_afu *afu = ctx->afu;
>>
>> @@ -710,7 +710,7 @@ void cxl_update_dedicated_ivtes_psl(struct
>> cxl_context *ctx)
>> ((u64)ctx->irqs.range[3] & 0xffff));
>> }
>>
>> -int cxl_attach_dedicated_process_psl(struct cxl_context *ctx, u64
>> wed, u64 amr)
>> +int cxl_attach_dedicated_process_psl8(struct cxl_context *ctx, u64
>> wed, u64 amr)
>> {
>> struct cxl_afu *afu = ctx->afu;
>> u64 pid;
>> @@ -880,7 +880,7 @@ static int native_get_irq_info(struct cxl_afu
>> *afu, struct cxl_irq_info *info)
>> return 0;
>> }
>>
>> -void cxl_native_irq_dump_regs_psl(struct cxl_context *ctx)
>> +void cxl_native_irq_dump_regs_psl8(struct cxl_context *ctx)
>> {
>> u64 fir1, fir2, fir_slice, serr, afu_debug;
>>
>> @@ -916,7 +916,7 @@ static irqreturn_t
>> native_handle_psl_slice_error(struct cxl_context *ctx,
>> return cxl_ops->ack_irq(ctx, 0, errstat);
>> }
>>
>> -irqreturn_t cxl_fail_irq_psl(struct cxl_afu *afu, struct
>> cxl_irq_info *irq_info)
>> +irqreturn_t cxl_fail_irq_psl8(struct cxl_afu *afu, struct
>> cxl_irq_info *irq_info)
>> {
>> if (irq_info->dsisr & CXL_PSL_DSISR_TRANS)
>> cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_AE);
>> diff --git a/drivers/misc/cxl/pci.c b/drivers/misc/cxl/pci.c
>> index 4d475a5..b952527 100644
>> --- a/drivers/misc/cxl/pci.c
>> +++ b/drivers/misc/cxl/pci.c
>> @@ -377,7 +377,7 @@ static int calc_capp_routing(struct pci_dev *dev,
>> u64 *chipid, u64 *capp_unit_id
>> return 0;
>> }
>>
>> -static int init_implementation_adapter_regs_psl(struct cxl *adapter,
>> struct pci_dev *dev)
>> +static int init_implementation_adapter_regs_psl8(struct cxl
>> *adapter, struct pci_dev *dev)
>> {
>> u64 psl_dsnctl, psl_fircntl;
>> u64 chipid;
>> @@ -434,7 +434,7 @@ static int
>> init_implementation_adapter_regs_xsl(struct cxl *adapter, struct pci_
>> /* For the PSL this is a multiple for 0 < n <= 7: */
>> #define PSL_2048_250MHZ_CYCLES 1
>>
>> -static void write_timebase_ctrl_psl(struct cxl *adapter)
>> +static void write_timebase_ctrl_psl8(struct cxl *adapter)
>> {
>> cxl_p1_write(adapter, CXL_PSL_TB_CTLSTAT,
>> TBSYNC_CNT(2 * PSL_2048_250MHZ_CYCLES));
>> @@ -455,7 +455,7 @@ static void write_timebase_ctrl_xsl(struct cxl
>> *adapter)
>> TBSYNC_CNT(XSL_4000_CLOCKS));
>> }
>>
>> -static u64 timebase_read_psl(struct cxl *adapter)
>> +static u64 timebase_read_psl8(struct cxl *adapter)
>> {
>> return cxl_p1_read(adapter, CXL_PSL_Timebase);
>> }
>> @@ -513,7 +513,7 @@ static void cxl_setup_psl_timebase(struct cxl
>> *adapter, struct pci_dev *dev)
>> return;
>> }
>>
>> -static int init_implementation_afu_regs_psl(struct cxl_afu *afu)
>> +static int init_implementation_afu_regs_psl8(struct cxl_afu *afu)
>> {
>> /* read/write masks for this slice */
>> cxl_p1n_write(afu, CXL_PSL_APCALLOC_A, 0xFFFFFFFEFEFEFEFEULL);
>> @@ -996,7 +996,7 @@ static int cxl_afu_descriptor_looks_ok(struct
>> cxl_afu *afu)
>> return 0;
>> }
>>
>> -static int sanitise_afu_regs_psl(struct cxl_afu *afu)
>> +static int sanitise_afu_regs_psl8(struct cxl_afu *afu)
>> {
>> u64 reg;
>>
>> @@ -1526,26 +1526,26 @@ static void cxl_deconfigure_adapter(struct
>> cxl *adapter)
>> pci_disable_device(pdev);
>> }
>>
>> -static const struct cxl_service_layer_ops psl_ops = {
>> - .adapter_regs_init = init_implementation_adapter_regs_psl,
>> - .invalidate_all = cxl_invalidate_all_psl,
>> - .afu_regs_init = init_implementation_afu_regs_psl,
>> - .sanitise_afu_regs = sanitise_afu_regs_psl,
>> +static const struct cxl_service_layer_ops psl8_ops = {
>> + .adapter_regs_init = init_implementation_adapter_regs_psl8,
>> + .invalidate_all = cxl_invalidate_all_psl8,
>> + .afu_regs_init = init_implementation_afu_regs_psl8,
>> + .sanitise_afu_regs = sanitise_afu_regs_psl8,
>> .register_serr_irq = cxl_native_register_serr_irq,
>> .release_serr_irq = cxl_native_release_serr_irq,
>> - .handle_interrupt = cxl_irq_psl,
>> - .fail_irq = cxl_fail_irq_psl,
>> - .activate_dedicated_process = cxl_activate_dedicated_process_psl,
>> - .attach_afu_directed = cxl_attach_afu_directed_psl,
>> - .attach_dedicated_process = cxl_attach_dedicated_process_psl,
>> - .update_dedicated_ivtes = cxl_update_dedicated_ivtes_psl,
>> - .debugfs_add_adapter_regs = cxl_debugfs_add_adapter_regs_psl,
>> - .debugfs_add_afu_regs = cxl_debugfs_add_afu_regs_psl,
>> - .psl_irq_dump_registers = cxl_native_irq_dump_regs_psl,
>> + .handle_interrupt = cxl_irq_psl8,
>> + .fail_irq = cxl_fail_irq_psl8,
>> + .activate_dedicated_process = cxl_activate_dedicated_process_psl8,
>> + .attach_afu_directed = cxl_attach_afu_directed_psl8,
>> + .attach_dedicated_process = cxl_attach_dedicated_process_psl8,
>> + .update_dedicated_ivtes = cxl_update_dedicated_ivtes_psl8,
>> + .debugfs_add_adapter_regs = cxl_debugfs_add_adapter_regs_psl8,
>> + .debugfs_add_afu_regs = cxl_debugfs_add_afu_regs_psl8,
>> + .psl_irq_dump_registers = cxl_native_irq_dump_regs_psl8,
>> .err_irq_dump_registers = cxl_native_err_irq_dump_regs,
>> - .debugfs_stop_trace = cxl_stop_trace_psl,
>> - .write_timebase_ctrl = write_timebase_ctrl_psl,
>> - .timebase_read = timebase_read_psl,
>> + .debugfs_stop_trace = cxl_stop_trace_psl8,
>> + .write_timebase_ctrl = write_timebase_ctrl_psl8,
>> + .timebase_read = timebase_read_psl8,
>> .capi_mode = OPAL_PHB_CAPI_MODE_CAPI,
>> .needs_reset_before_disable = true,
>> };
>> @@ -1566,8 +1566,8 @@ static void set_sl_ops(struct cxl *adapter,
>> struct pci_dev *dev)
>> adapter->native->sl_ops = &xsl_ops;
>> adapter->min_pe = 1; /* Workaround for CX-4 hardware bug */
>> } else {
>> - dev_info(&dev->dev, "Device uses a PSL\n");
>> - adapter->native->sl_ops = &psl_ops;
>> + dev_info(&dev->dev, "Device uses a PSL8\n");
>> + adapter->native->sl_ops = &psl8_ops;
>> }
>> }
>>
^ permalink raw reply [flat|nested] 20+ messages in thread
* [PATCH V2 5/6] cxl: Isolate few psl8 specific calls
2017-03-14 11:08 [PATCH V2 0/6] cxl: Add support for Coherent Accelerator Interface Architecture 2.0 Christophe Lombard
` (3 preceding siblings ...)
2017-03-14 11:08 ` [PATCH V2 4/6] cxl: Rename some psl8 specific functions Christophe Lombard
@ 2017-03-14 11:08 ` Christophe Lombard
2017-03-20 18:15 ` Frederic Barrat
2017-03-14 11:08 ` [PATCH V2 6/6] cxl: Add psl9 specific code Christophe Lombard
2017-03-21 2:47 ` [PATCH V2 0/6] cxl: Add support for Coherent Accelerator Interface Architecture 2.0 Andrew Donnellan
6 siblings, 1 reply; 20+ messages in thread
From: Christophe Lombard @ 2017-03-14 11:08 UTC (permalink / raw)
To: linuxppc-dev, fbarrat, imunsie, andrew.donnellan
Point out the specific Coherent Accelerator Interface Architecture,
level 1, registers.
Code and functions specific to PSL8 (CAIA1) must be framed.
Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com>
---
drivers/misc/cxl/context.c | 28 +++++++++++---------
drivers/misc/cxl/cxl.h | 35 +++++++++++++++++++------
drivers/misc/cxl/debugfs.c | 6 +++--
drivers/misc/cxl/fault.c | 14 +++++-----
drivers/misc/cxl/native.c | 58 ++++++++++++++++++++++++++---------------
drivers/misc/cxl/pci.c | 64 +++++++++++++++++++++++++++++++---------------
6 files changed, 136 insertions(+), 69 deletions(-)
diff --git a/drivers/misc/cxl/context.c b/drivers/misc/cxl/context.c
index ed0a447..0fd3cf8 100644
--- a/drivers/misc/cxl/context.c
+++ b/drivers/misc/cxl/context.c
@@ -39,23 +39,26 @@ int cxl_context_init(struct cxl_context *ctx, struct cxl_afu *afu, bool master)
{
int i;
- spin_lock_init(&ctx->sste_lock);
+ if (cxl_is_psl8(afu))
+ spin_lock_init(&ctx->sste_lock);
ctx->afu = afu;
ctx->master = master;
ctx->pid = NULL; /* Set in start work ioctl */
mutex_init(&ctx->mapping_lock);
ctx->mapping = NULL;
- /*
- * Allocate the segment table before we put it in the IDR so that we
- * can always access it when dereferenced from IDR. For the same
- * reason, the segment table is only destroyed after the context is
- * removed from the IDR. Access to this in the IOCTL is protected by
- * Linux filesytem symantics (can't IOCTL until open is complete).
- */
- i = cxl_alloc_sst(ctx);
- if (i)
- return i;
+ if (cxl_is_psl8(afu)) {
+ /*
+ * Allocate the segment table before we put it in the IDR so that we
+ * can always access it when dereferenced from IDR. For the same
+ * reason, the segment table is only destroyed after the context is
+ * removed from the IDR. Access to this in the IOCTL is protected by
+ * Linux filesytem symantics (can't IOCTL until open is complete).
+ */
+ i = cxl_alloc_sst(ctx);
+ if (i)
+ return i;
+ }
INIT_WORK(&ctx->fault_work, cxl_handle_fault);
@@ -307,7 +310,8 @@ static void reclaim_ctx(struct rcu_head *rcu)
{
struct cxl_context *ctx = container_of(rcu, struct cxl_context, rcu);
- free_page((u64)ctx->sstp);
+ if (cxl_is_psl8(ctx->afu))
+ free_page((u64)ctx->sstp);
if (ctx->ff_page)
__free_page(ctx->ff_page);
ctx->sstp = NULL;
diff --git a/drivers/misc/cxl/cxl.h b/drivers/misc/cxl/cxl.h
index 1f04238..f6a3a34 100644
--- a/drivers/misc/cxl/cxl.h
+++ b/drivers/misc/cxl/cxl.h
@@ -73,7 +73,7 @@ static const cxl_p1_reg_t CXL_PSL_Control = {0x0020};
static const cxl_p1_reg_t CXL_PSL_DLCNTL = {0x0060};
static const cxl_p1_reg_t CXL_PSL_DLADDR = {0x0068};
-/* PSL Lookaside Buffer Management Area */
+/* PSL Lookaside Buffer Management Area - CAIA 1 */
static const cxl_p1_reg_t CXL_PSL_LBISEL = {0x0080};
static const cxl_p1_reg_t CXL_PSL_SLBIE = {0x0088};
static const cxl_p1_reg_t CXL_PSL_SLBIA = {0x0090};
@@ -82,7 +82,7 @@ static const cxl_p1_reg_t CXL_PSL_TLBIA = {0x00A8};
static const cxl_p1_reg_t CXL_PSL_AFUSEL = {0x00B0};
/* 0x00C0:7EFF Implementation dependent area */
-/* PSL registers */
+/* PSL registers - CAIA 1 */
static const cxl_p1_reg_t CXL_PSL_FIR1 = {0x0100};
static const cxl_p1_reg_t CXL_PSL_FIR2 = {0x0108};
static const cxl_p1_reg_t CXL_PSL_Timebase = {0x0110};
@@ -109,7 +109,7 @@ static const cxl_p1n_reg_t CXL_PSL_AMBAR_An = {0x10};
static const cxl_p1n_reg_t CXL_PSL_SPOffset_An = {0x18};
static const cxl_p1n_reg_t CXL_PSL_ID_An = {0x20};
static const cxl_p1n_reg_t CXL_PSL_SERR_An = {0x28};
-/* Memory Management and Lookaside Buffer Management */
+/* Memory Management and Lookaside Buffer Management - CAIA 1*/
static const cxl_p1n_reg_t CXL_PSL_SDR_An = {0x30};
static const cxl_p1n_reg_t CXL_PSL_AMOR_An = {0x38};
/* Pointer Area */
@@ -124,6 +124,7 @@ static const cxl_p1n_reg_t CXL_PSL_IVTE_Limit_An = {0xB8};
/* 0xC0:FF Implementation Dependent Area */
static const cxl_p1n_reg_t CXL_PSL_FIR_SLICE_An = {0xC0};
static const cxl_p1n_reg_t CXL_AFU_DEBUG_An = {0xC8};
+/* 0xC0:FF Implementation Dependent Area - CAIA 1 */
static const cxl_p1n_reg_t CXL_PSL_APCALLOC_A = {0xD0};
static const cxl_p1n_reg_t CXL_PSL_COALLOC_A = {0xD8};
static const cxl_p1n_reg_t CXL_PSL_RXCTL_A = {0xE0};
@@ -133,12 +134,14 @@ static const cxl_p1n_reg_t CXL_PSL_SLICE_TRACE = {0xE8};
/* Configuration and Control Area */
static const cxl_p2n_reg_t CXL_PSL_PID_TID_An = {0x000};
static const cxl_p2n_reg_t CXL_CSRP_An = {0x008};
+/* Configuration and Control Area - CAIA 1 */
static const cxl_p2n_reg_t CXL_AURP0_An = {0x010};
static const cxl_p2n_reg_t CXL_AURP1_An = {0x018};
static const cxl_p2n_reg_t CXL_SSTP0_An = {0x020};
static const cxl_p2n_reg_t CXL_SSTP1_An = {0x028};
+/* Configuration and Control Area - CAIA 1 */
static const cxl_p2n_reg_t CXL_PSL_AMR_An = {0x030};
-/* Segment Lookaside Buffer Management */
+/* Segment Lookaside Buffer Management - CAIA 1 */
static const cxl_p2n_reg_t CXL_SLBIE_An = {0x040};
static const cxl_p2n_reg_t CXL_SLBIA_An = {0x048};
static const cxl_p2n_reg_t CXL_SLBI_Select_An = {0x050};
@@ -257,7 +260,7 @@ static const cxl_p2n_reg_t CXL_PSL_WED_An = {0x0A0};
#define CXL_SSTP1_An_STVA_L_MASK (~((1ull << (63-55))-1))
#define CXL_SSTP1_An_V (1ull << (63-63))
-/****** CXL_PSL_SLBIE_[An] **************************************************/
+/****** CXL_PSL_SLBIE_[An] - CAIA 1 **************************************************/
/* write: */
#define CXL_SLBIE_C PPC_BIT(36) /* Class */
#define CXL_SLBIE_SS PPC_BITMASK(37, 38) /* Segment Size */
@@ -267,10 +270,10 @@ static const cxl_p2n_reg_t CXL_PSL_WED_An = {0x0A0};
#define CXL_SLBIE_MAX PPC_BITMASK(24, 31)
#define CXL_SLBIE_PENDING PPC_BITMASK(56, 63)
-/****** Common to all CXL_TLBIA/SLBIA_[An] **********************************/
+/****** Common to all CXL_TLBIA/SLBIA_[An] - CAIA 1 **********************************/
#define CXL_TLB_SLB_P (1ull) /* Pending (read) */
-/****** Common to all CXL_TLB/SLB_IA/IE_[An] registers **********************/
+/****** Common to all CXL_TLB/SLB_IA/IE_[An] registers - CAIA 1 **********************/
#define CXL_TLB_SLB_IQ_ALL (0ull) /* Inv qualifier */
#define CXL_TLB_SLB_IQ_LPID (1ull) /* Inv qualifier */
#define CXL_TLB_SLB_IQ_LPIDPID (3ull) /* Inv qualifier */
@@ -278,7 +281,7 @@ static const cxl_p2n_reg_t CXL_PSL_WED_An = {0x0A0};
/****** CXL_PSL_AFUSEL ******************************************************/
#define CXL_PSL_AFUSEL_A (1ull << (63-55)) /* Adapter wide invalidates affect all AFUs */
-/****** CXL_PSL_DSISR_An ****************************************************/
+/****** CXL_PSL_DSISR_An - CAIA 1 ****************************************************/
#define CXL_PSL_DSISR_An_DS (1ull << (63-0)) /* Segment not found */
#define CXL_PSL_DSISR_An_DM (1ull << (63-1)) /* PTE not found (See also: M) or protection fault */
#define CXL_PSL_DSISR_An_ST (1ull << (63-2)) /* Segment Table PTE not found */
@@ -749,6 +752,22 @@ static inline u64 cxl_p2n_read(struct cxl_afu *afu, cxl_p2n_reg_t reg)
return ~0ULL;
}
+static inline bool cxl_is_power8(void)
+{
+ if ((pvr_version_is(PVR_POWER8E)) ||
+ (pvr_version_is(PVR_POWER8NVL)) ||
+ (pvr_version_is(PVR_POWER8)))
+ return true;
+ return false;
+}
+
+static inline bool cxl_is_psl8(struct cxl_afu *afu)
+{
+ if (afu->adapter->caia_major == 1)
+ return true;
+ return false;
+}
+
ssize_t cxl_pci_afu_read_err_buffer(struct cxl_afu *afu, char *buf,
loff_t off, size_t count);
diff --git a/drivers/misc/cxl/debugfs.c b/drivers/misc/cxl/debugfs.c
index 2ff10a9..43a1a27 100644
--- a/drivers/misc/cxl/debugfs.c
+++ b/drivers/misc/cxl/debugfs.c
@@ -94,6 +94,9 @@ void cxl_debugfs_adapter_remove(struct cxl *adapter)
void cxl_debugfs_add_afu_regs_psl8(struct cxl_afu *afu, struct dentry *dir)
{
+ debugfs_create_io_x64("sstp0", S_IRUSR, dir, _cxl_p2n_addr(afu, CXL_SSTP0_An));
+ debugfs_create_io_x64("sstp1", S_IRUSR, dir, _cxl_p2n_addr(afu, CXL_SSTP1_An));
+
debugfs_create_io_x64("fir", S_IRUSR, dir, _cxl_p1n_addr(afu, CXL_PSL_FIR_SLICE_An));
debugfs_create_io_x64("serr", S_IRUSR, dir, _cxl_p1n_addr(afu, CXL_PSL_SERR_An));
debugfs_create_io_x64("afu_debug", S_IRUSR, dir, _cxl_p1n_addr(afu, CXL_AFU_DEBUG_An));
@@ -117,8 +120,7 @@ int cxl_debugfs_afu_add(struct cxl_afu *afu)
debugfs_create_io_x64("sr", S_IRUSR, dir, _cxl_p1n_addr(afu, CXL_PSL_SR_An));
debugfs_create_io_x64("dsisr", S_IRUSR, dir, _cxl_p2n_addr(afu, CXL_PSL_DSISR_An));
debugfs_create_io_x64("dar", S_IRUSR, dir, _cxl_p2n_addr(afu, CXL_PSL_DAR_An));
- debugfs_create_io_x64("sstp0", S_IRUSR, dir, _cxl_p2n_addr(afu, CXL_SSTP0_An));
- debugfs_create_io_x64("sstp1", S_IRUSR, dir, _cxl_p2n_addr(afu, CXL_SSTP1_An));
+
debugfs_create_io_x64("err_status", S_IRUSR, dir, _cxl_p2n_addr(afu, CXL_PSL_ErrStat_An));
if (afu->adapter->native->sl_ops->debugfs_add_afu_regs)
diff --git a/drivers/misc/cxl/fault.c b/drivers/misc/cxl/fault.c
index 14a5bfa..9b72b90 100644
--- a/drivers/misc/cxl/fault.c
+++ b/drivers/misc/cxl/fault.c
@@ -231,12 +231,14 @@ void cxl_handle_fault(struct work_struct *fault_work)
}
}
- if (dsisr & CXL_PSL_DSISR_An_DS)
- cxl_handle_segment_miss(ctx, mm, dar);
- else if (dsisr & CXL_PSL_DSISR_An_DM)
- cxl_handle_page_fault(ctx, mm, dsisr, dar);
- else
- WARN(1, "cxl_handle_fault has nothing to handle\n");
+ if (cxl_is_psl8(ctx->afu)) {
+ if (dsisr & CXL_PSL_DSISR_An_DS)
+ cxl_handle_segment_miss(ctx, mm, dar);
+ else if (dsisr & CXL_PSL_DSISR_An_DM)
+ cxl_handle_page_fault(ctx, mm, dsisr, dar);
+ else
+ WARN(1, "cxl_handle_fault has nothing to handle\n");
+ }
if (mm)
mmput(mm);
diff --git a/drivers/misc/cxl/native.c b/drivers/misc/cxl/native.c
index ea08605..51aefcb 100644
--- a/drivers/misc/cxl/native.c
+++ b/drivers/misc/cxl/native.c
@@ -155,16 +155,25 @@ int cxl_psl_purge(struct cxl_afu *afu)
}
dsisr = cxl_p2n_read(afu, CXL_PSL_DSISR_An);
- pr_devel_ratelimited("PSL purging... PSL_CNTL: 0x%016llx PSL_DSISR: 0x%016llx\n", PSL_CNTL, dsisr);
- if (dsisr & CXL_PSL_DSISR_TRANS) {
- dar = cxl_p2n_read(afu, CXL_PSL_DAR_An);
- dev_notice(&afu->dev, "PSL purge terminating pending translation, DSISR: 0x%016llx, DAR: 0x%016llx\n", dsisr, dar);
- cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_AE);
- } else if (dsisr) {
- dev_notice(&afu->dev, "PSL purge acknowledging pending non-translation fault, DSISR: 0x%016llx\n", dsisr);
- cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_A);
- } else {
- cpu_relax();
+ pr_devel_ratelimited("PSL purging... PSL_CNTL: 0x%016llx"
+ " PSL_DSISR: 0x%016llx\n",
+ PSL_CNTL, dsisr);
+ if (cxl_is_psl8(afu)) {
+ if (dsisr & CXL_PSL_DSISR_TRANS) {
+ dar = cxl_p2n_read(afu, CXL_PSL_DAR_An);
+ dev_notice(&afu->dev, "PSL purge terminating "
+ "pending translation, "
+ "DSISR: 0x%016llx, DAR: 0x%016llx\n",
+ dsisr, dar);
+ cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_AE);
+ } else if (dsisr) {
+ dev_notice(&afu->dev, "PSL purge acknowledging "
+ "pending non-translation fault, "
+ "DSISR: 0x%016llx\n", dsisr);
+ cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_A);
+ } else {
+ cpu_relax();
+ }
}
PSL_CNTL = cxl_p1n_read(afu, CXL_PSL_SCNTL_An);
}
@@ -466,7 +475,8 @@ static int remove_process_element(struct cxl_context *ctx)
if (!rc)
ctx->pe_inserted = false;
- slb_invalid(ctx);
+ if (cxl_is_power8())
+ slb_invalid(ctx);
pr_devel("%s Remove pe: %i finished\n", __func__, ctx->pe);
mutex_unlock(&ctx->afu->native->spa_mutex);
@@ -499,7 +509,8 @@ static int activate_afu_directed(struct cxl_afu *afu)
attach_spa(afu);
cxl_p1n_write(afu, CXL_PSL_SCNTL_An, CXL_PSL_SCNTL_An_PM_AFU);
- cxl_p1n_write(afu, CXL_PSL_AMOR_An, 0xFFFFFFFFFFFFFFFFULL);
+ if (cxl_is_power8())
+ cxl_p1n_write(afu, CXL_PSL_AMOR_An, 0xFFFFFFFFFFFFFFFFULL);
cxl_p1n_write(afu, CXL_PSL_ID_An, CXL_PSL_ID_An_F | CXL_PSL_ID_An_L);
afu->current_mode = CXL_MODE_DIRECTED;
@@ -872,7 +883,8 @@ static int native_get_irq_info(struct cxl_afu *afu, struct cxl_irq_info *info)
info->dsisr = cxl_p2n_read(afu, CXL_PSL_DSISR_An);
info->dar = cxl_p2n_read(afu, CXL_PSL_DAR_An);
- info->dsr = cxl_p2n_read(afu, CXL_PSL_DSR_An);
+ if (cxl_is_power8())
+ info->dsr = cxl_p2n_read(afu, CXL_PSL_DSR_An);
info->afu_err = cxl_p2n_read(afu, CXL_AFU_ERR_An);
info->errstat = cxl_p2n_read(afu, CXL_PSL_ErrStat_An);
info->proc_handle = 0;
@@ -984,7 +996,8 @@ static void native_irq_wait(struct cxl_context *ctx)
if (ph != ctx->pe)
return;
dsisr = cxl_p2n_read(ctx->afu, CXL_PSL_DSISR_An);
- if ((dsisr & CXL_PSL_DSISR_PENDING) == 0)
+ if (cxl_is_psl8(ctx->afu) &&
+ ((dsisr & CXL_PSL_DSISR_PENDING) == 0))
return;
/*
* We are waiting for the workqueue to process our
@@ -1001,21 +1014,25 @@ static void native_irq_wait(struct cxl_context *ctx)
static irqreturn_t native_slice_irq_err(int irq, void *data)
{
struct cxl_afu *afu = data;
- u64 fir_slice, errstat, serr, afu_debug, afu_error, dsisr;
+ u64 errstat, serr, afu_error, dsisr;
+ u64 fir_slice, afu_debug;
/*
* slice err interrupt is only used with full PSL (no XSL)
*/
serr = cxl_p1n_read(afu, CXL_PSL_SERR_An);
- fir_slice = cxl_p1n_read(afu, CXL_PSL_FIR_SLICE_An);
errstat = cxl_p2n_read(afu, CXL_PSL_ErrStat_An);
- afu_debug = cxl_p1n_read(afu, CXL_AFU_DEBUG_An);
afu_error = cxl_p2n_read(afu, CXL_AFU_ERR_An);
dsisr = cxl_p2n_read(afu, CXL_PSL_DSISR_An);
cxl_afu_decode_psl_serr(afu, serr);
- dev_crit(&afu->dev, "PSL_FIR_SLICE_An: 0x%016llx\n", fir_slice);
+
+ if (cxl_is_power8()) {
+ fir_slice = cxl_p1n_read(afu, CXL_PSL_FIR_SLICE_An);
+ afu_debug = cxl_p1n_read(afu, CXL_AFU_DEBUG_An);
+ dev_crit(&afu->dev, "PSL_FIR_SLICE_An: 0x%016llx\n", fir_slice);
+ dev_crit(&afu->dev, "CXL_PSL_AFU_DEBUG_An: 0x%016llx\n", afu_debug);
+ }
dev_crit(&afu->dev, "CXL_PSL_ErrStat_An: 0x%016llx\n", errstat);
- dev_crit(&afu->dev, "CXL_PSL_AFU_DEBUG_An: 0x%016llx\n", afu_debug);
dev_crit(&afu->dev, "AFU_ERR_An: 0x%.16llx\n", afu_error);
dev_crit(&afu->dev, "PSL_DSISR_An: 0x%.16llx\n", dsisr);
@@ -1108,7 +1125,8 @@ int cxl_native_register_serr_irq(struct cxl_afu *afu)
}
serr = cxl_p1n_read(afu, CXL_PSL_SERR_An);
- serr = (serr & 0x00ffffffffff0000ULL) | (afu->serr_hwirq & 0xffff);
+ if (cxl_is_power8())
+ serr = (serr & 0x00ffffffffff0000ULL) | (afu->serr_hwirq & 0xffff);
cxl_p1n_write(afu, CXL_PSL_SERR_An, serr);
return 0;
diff --git a/drivers/misc/cxl/pci.c b/drivers/misc/cxl/pci.c
index b952527..5880092 100644
--- a/drivers/misc/cxl/pci.c
+++ b/drivers/misc/cxl/pci.c
@@ -324,32 +324,33 @@ static void dump_afu_descriptor(struct cxl_afu *afu)
#undef show_reg
}
-#define CAPP_UNIT0_ID 0xBA
-#define CAPP_UNIT1_ID 0XBE
+#define P8_CAPP_UNIT0_ID 0xBA
+#define P8_CAPP_UNIT1_ID 0XBE
static u64 get_capp_unit_id(struct device_node *np)
{
u32 phb_index;
- /*
- * For chips other than POWER8NVL, we only have CAPP 0,
- * irrespective of which PHB is used.
- */
- if (!pvr_version_is(PVR_POWER8NVL))
- return CAPP_UNIT0_ID;
+ if (of_property_read_u32(np, "ibm,phb-index", &phb_index))
+ return 0;
/*
- * For POWER8NVL, assume CAPP 0 is attached to PHB0 and
- * CAPP 1 is attached to PHB1.
+ * POWER 8:
+ * - For chips other than POWER8NVL, we only have CAPP 0,
+ * irrespective of which PHB is used.
+ * - For POWER8NVL, assume CAPP 0 is attached to PHB0 and
+ * CAPP 1 is attached to PHB1.
*/
- if (of_property_read_u32(np, "ibm,phb-index", &phb_index))
- return 0;
+ if (cxl_is_power8()) {
+ if (!pvr_version_is(PVR_POWER8NVL))
+ return P8_CAPP_UNIT0_ID;
- if (phb_index == 0)
- return CAPP_UNIT0_ID;
+ if (phb_index == 0)
+ return P8_CAPP_UNIT0_ID;
- if (phb_index == 1)
- return CAPP_UNIT1_ID;
+ if (phb_index == 1)
+ return P8_CAPP_UNIT1_ID;
+ }
return 0;
}
@@ -968,7 +969,7 @@ static int cxl_afu_descriptor_looks_ok(struct cxl_afu *afu)
}
if (afu->pp_psa && (afu->pp_size < PAGE_SIZE))
- dev_warn(&afu->dev, "AFU uses < PAGE_SIZE per-process PSA!");
+ dev_warn(&afu->dev, "AFU uses pp_size(%#016llx) < PAGE_SIZE per-process PSA!\n", afu->pp_size);
for (i = 0; i < afu->crs_num; i++) {
rc = cxl_ops->afu_cr_read32(afu, i, 0, &val);
@@ -1251,8 +1252,13 @@ int cxl_pci_reset(struct cxl *adapter)
dev_info(&dev->dev, "CXL reset\n");
- /* the adapter is about to be reset, so ignore errors */
- cxl_data_cache_flush(adapter);
+ /*
+ * The adapter is about to be reset, so ignore errors.
+ * Not supported on P9 DD1 but don't forget to enable it
+ * on P9 DD2
+ */
+ if (cxl_is_power8())
+ cxl_data_cache_flush(adapter);
/* pcie_warm_reset requests a fundamental pci reset which includes a
* PERST assert/deassert. PERST triggers a loading of the image
@@ -1382,6 +1388,14 @@ static void cxl_fixup_malformed_tlp(struct cxl *adapter, struct pci_dev *dev)
pci_write_config_dword(dev, aer + PCI_ERR_UNCOR_MASK, data);
}
+static bool cxl_compatible_caia_version(struct cxl *adapter)
+{
+ if (cxl_is_power8() && (adapter->caia_major == 1))
+ return true;
+
+ return false;
+}
+
static int cxl_vsec_looks_ok(struct cxl *adapter, struct pci_dev *dev)
{
if (adapter->vsec_status & CXL_STATUS_SECOND_PORT)
@@ -1392,6 +1406,12 @@ static int cxl_vsec_looks_ok(struct cxl *adapter, struct pci_dev *dev)
return -EINVAL;
}
+ if (!cxl_compatible_caia_version(adapter)) {
+ dev_info(&dev->dev, "Ignoring card. PSL type is not supported "
+ "(caia version: %d)\n", adapter->caia_major);
+ return -ENODEV;
+ }
+
if (!adapter->slices) {
/* Once we support dynamic reprogramming we can use the card if
* it supports loadable AFUs */
@@ -1566,8 +1586,10 @@ static void set_sl_ops(struct cxl *adapter, struct pci_dev *dev)
adapter->native->sl_ops = &xsl_ops;
adapter->min_pe = 1; /* Workaround for CX-4 hardware bug */
} else {
- dev_info(&dev->dev, "Device uses a PSL8\n");
- adapter->native->sl_ops = &psl8_ops;
+ if (cxl_is_power8()) {
+ dev_info(&dev->dev, "Device uses a PSL8\n");
+ adapter->native->sl_ops = &psl8_ops;
+ }
}
}
--
2.7.4
^ permalink raw reply related [flat|nested] 20+ messages in thread* Re: [PATCH V2 5/6] cxl: Isolate few psl8 specific calls
2017-03-14 11:08 ` [PATCH V2 5/6] cxl: Isolate few psl8 specific calls Christophe Lombard
@ 2017-03-20 18:15 ` Frederic Barrat
0 siblings, 0 replies; 20+ messages in thread
From: Frederic Barrat @ 2017-03-20 18:15 UTC (permalink / raw)
To: Christophe Lombard, linuxppc-dev, imunsie, andrew.donnellan
Le 14/03/2017 à 12:08, Christophe Lombard a écrit :
> Point out the specific Coherent Accelerator Interface Architecture,
> level 1, registers.
> Code and functions specific to PSL8 (CAIA1) must be framed.
>
> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com>
> ---
> drivers/misc/cxl/context.c | 28 +++++++++++---------
> drivers/misc/cxl/cxl.h | 35 +++++++++++++++++++------
> drivers/misc/cxl/debugfs.c | 6 +++--
> drivers/misc/cxl/fault.c | 14 +++++-----
> drivers/misc/cxl/native.c | 58 ++++++++++++++++++++++++++---------------
> drivers/misc/cxl/pci.c | 64 +++++++++++++++++++++++++++++++---------------
> 6 files changed, 136 insertions(+), 69 deletions(-)
>
> diff --git a/drivers/misc/cxl/context.c b/drivers/misc/cxl/context.c
> index ed0a447..0fd3cf8 100644
> --- a/drivers/misc/cxl/context.c
> +++ b/drivers/misc/cxl/context.c
> @@ -39,23 +39,26 @@ int cxl_context_init(struct cxl_context *ctx, struct cxl_afu *afu, bool master)
> {
> int i;
>
> - spin_lock_init(&ctx->sste_lock);
> + if (cxl_is_psl8(afu))
> + spin_lock_init(&ctx->sste_lock);
> ctx->afu = afu;
> ctx->master = master;
> ctx->pid = NULL; /* Set in start work ioctl */
> mutex_init(&ctx->mapping_lock);
> ctx->mapping = NULL;
>
> - /*
> - * Allocate the segment table before we put it in the IDR so that we
> - * can always access it when dereferenced from IDR. For the same
> - * reason, the segment table is only destroyed after the context is
> - * removed from the IDR. Access to this in the IOCTL is protected by
> - * Linux filesytem symantics (can't IOCTL until open is complete).
> - */
> - i = cxl_alloc_sst(ctx);
> - if (i)
> - return i;
> + if (cxl_is_psl8(afu)) {
> + /*
> + * Allocate the segment table before we put it in the IDR so that we
> + * can always access it when dereferenced from IDR. For the same
> + * reason, the segment table is only destroyed after the context is
> + * removed from the IDR. Access to this in the IOCTL is protected by
> + * Linux filesytem symantics (can't IOCTL until open is complete).
> + */
> + i = cxl_alloc_sst(ctx);
> + if (i)
> + return i;
> + }
I guess we could have a subfunction context_init_psl8() to do the
psl8-related initializations, but since there's nothing specific to p9
and the psl8-only code is quite small, I won't fight it. We should
probably regroup the sste_lock and sst alloc under the same "if"
statement though.
The rest of the patch looks ok.
Fred
> INIT_WORK(&ctx->fault_work, cxl_handle_fault);
>
> @@ -307,7 +310,8 @@ static void reclaim_ctx(struct rcu_head *rcu)
> {
> struct cxl_context *ctx = container_of(rcu, struct cxl_context, rcu);
>
> - free_page((u64)ctx->sstp);
> + if (cxl_is_psl8(ctx->afu))
> + free_page((u64)ctx->sstp);
> if (ctx->ff_page)
> __free_page(ctx->ff_page);
> ctx->sstp = NULL;
> diff --git a/drivers/misc/cxl/cxl.h b/drivers/misc/cxl/cxl.h
> index 1f04238..f6a3a34 100644
> --- a/drivers/misc/cxl/cxl.h
> +++ b/drivers/misc/cxl/cxl.h
> @@ -73,7 +73,7 @@ static const cxl_p1_reg_t CXL_PSL_Control = {0x0020};
> static const cxl_p1_reg_t CXL_PSL_DLCNTL = {0x0060};
> static const cxl_p1_reg_t CXL_PSL_DLADDR = {0x0068};
>
> -/* PSL Lookaside Buffer Management Area */
> +/* PSL Lookaside Buffer Management Area - CAIA 1 */
> static const cxl_p1_reg_t CXL_PSL_LBISEL = {0x0080};
> static const cxl_p1_reg_t CXL_PSL_SLBIE = {0x0088};
> static const cxl_p1_reg_t CXL_PSL_SLBIA = {0x0090};
> @@ -82,7 +82,7 @@ static const cxl_p1_reg_t CXL_PSL_TLBIA = {0x00A8};
> static const cxl_p1_reg_t CXL_PSL_AFUSEL = {0x00B0};
>
> /* 0x00C0:7EFF Implementation dependent area */
> -/* PSL registers */
> +/* PSL registers - CAIA 1 */
> static const cxl_p1_reg_t CXL_PSL_FIR1 = {0x0100};
> static const cxl_p1_reg_t CXL_PSL_FIR2 = {0x0108};
> static const cxl_p1_reg_t CXL_PSL_Timebase = {0x0110};
> @@ -109,7 +109,7 @@ static const cxl_p1n_reg_t CXL_PSL_AMBAR_An = {0x10};
> static const cxl_p1n_reg_t CXL_PSL_SPOffset_An = {0x18};
> static const cxl_p1n_reg_t CXL_PSL_ID_An = {0x20};
> static const cxl_p1n_reg_t CXL_PSL_SERR_An = {0x28};
> -/* Memory Management and Lookaside Buffer Management */
> +/* Memory Management and Lookaside Buffer Management - CAIA 1*/
> static const cxl_p1n_reg_t CXL_PSL_SDR_An = {0x30};
> static const cxl_p1n_reg_t CXL_PSL_AMOR_An = {0x38};
> /* Pointer Area */
> @@ -124,6 +124,7 @@ static const cxl_p1n_reg_t CXL_PSL_IVTE_Limit_An = {0xB8};
> /* 0xC0:FF Implementation Dependent Area */
> static const cxl_p1n_reg_t CXL_PSL_FIR_SLICE_An = {0xC0};
> static const cxl_p1n_reg_t CXL_AFU_DEBUG_An = {0xC8};
> +/* 0xC0:FF Implementation Dependent Area - CAIA 1 */
> static const cxl_p1n_reg_t CXL_PSL_APCALLOC_A = {0xD0};
> static const cxl_p1n_reg_t CXL_PSL_COALLOC_A = {0xD8};
> static const cxl_p1n_reg_t CXL_PSL_RXCTL_A = {0xE0};
> @@ -133,12 +134,14 @@ static const cxl_p1n_reg_t CXL_PSL_SLICE_TRACE = {0xE8};
> /* Configuration and Control Area */
> static const cxl_p2n_reg_t CXL_PSL_PID_TID_An = {0x000};
> static const cxl_p2n_reg_t CXL_CSRP_An = {0x008};
> +/* Configuration and Control Area - CAIA 1 */
> static const cxl_p2n_reg_t CXL_AURP0_An = {0x010};
> static const cxl_p2n_reg_t CXL_AURP1_An = {0x018};
> static const cxl_p2n_reg_t CXL_SSTP0_An = {0x020};
> static const cxl_p2n_reg_t CXL_SSTP1_An = {0x028};
> +/* Configuration and Control Area - CAIA 1 */
> static const cxl_p2n_reg_t CXL_PSL_AMR_An = {0x030};
> -/* Segment Lookaside Buffer Management */
> +/* Segment Lookaside Buffer Management - CAIA 1 */
> static const cxl_p2n_reg_t CXL_SLBIE_An = {0x040};
> static const cxl_p2n_reg_t CXL_SLBIA_An = {0x048};
> static const cxl_p2n_reg_t CXL_SLBI_Select_An = {0x050};
> @@ -257,7 +260,7 @@ static const cxl_p2n_reg_t CXL_PSL_WED_An = {0x0A0};
> #define CXL_SSTP1_An_STVA_L_MASK (~((1ull << (63-55))-1))
> #define CXL_SSTP1_An_V (1ull << (63-63))
>
> -/****** CXL_PSL_SLBIE_[An] **************************************************/
> +/****** CXL_PSL_SLBIE_[An] - CAIA 1 **************************************************/
> /* write: */
> #define CXL_SLBIE_C PPC_BIT(36) /* Class */
> #define CXL_SLBIE_SS PPC_BITMASK(37, 38) /* Segment Size */
> @@ -267,10 +270,10 @@ static const cxl_p2n_reg_t CXL_PSL_WED_An = {0x0A0};
> #define CXL_SLBIE_MAX PPC_BITMASK(24, 31)
> #define CXL_SLBIE_PENDING PPC_BITMASK(56, 63)
>
> -/****** Common to all CXL_TLBIA/SLBIA_[An] **********************************/
> +/****** Common to all CXL_TLBIA/SLBIA_[An] - CAIA 1 **********************************/
> #define CXL_TLB_SLB_P (1ull) /* Pending (read) */
>
> -/****** Common to all CXL_TLB/SLB_IA/IE_[An] registers **********************/
> +/****** Common to all CXL_TLB/SLB_IA/IE_[An] registers - CAIA 1 **********************/
> #define CXL_TLB_SLB_IQ_ALL (0ull) /* Inv qualifier */
> #define CXL_TLB_SLB_IQ_LPID (1ull) /* Inv qualifier */
> #define CXL_TLB_SLB_IQ_LPIDPID (3ull) /* Inv qualifier */
> @@ -278,7 +281,7 @@ static const cxl_p2n_reg_t CXL_PSL_WED_An = {0x0A0};
> /****** CXL_PSL_AFUSEL ******************************************************/
> #define CXL_PSL_AFUSEL_A (1ull << (63-55)) /* Adapter wide invalidates affect all AFUs */
>
> -/****** CXL_PSL_DSISR_An ****************************************************/
> +/****** CXL_PSL_DSISR_An - CAIA 1 ****************************************************/
> #define CXL_PSL_DSISR_An_DS (1ull << (63-0)) /* Segment not found */
> #define CXL_PSL_DSISR_An_DM (1ull << (63-1)) /* PTE not found (See also: M) or protection fault */
> #define CXL_PSL_DSISR_An_ST (1ull << (63-2)) /* Segment Table PTE not found */
> @@ -749,6 +752,22 @@ static inline u64 cxl_p2n_read(struct cxl_afu *afu, cxl_p2n_reg_t reg)
> return ~0ULL;
> }
>
> +static inline bool cxl_is_power8(void)
> +{
> + if ((pvr_version_is(PVR_POWER8E)) ||
> + (pvr_version_is(PVR_POWER8NVL)) ||
> + (pvr_version_is(PVR_POWER8)))
> + return true;
> + return false;
> +}
> +
> +static inline bool cxl_is_psl8(struct cxl_afu *afu)
> +{
> + if (afu->adapter->caia_major == 1)
> + return true;
> + return false;
> +}
> +
> ssize_t cxl_pci_afu_read_err_buffer(struct cxl_afu *afu, char *buf,
> loff_t off, size_t count);
>
> diff --git a/drivers/misc/cxl/debugfs.c b/drivers/misc/cxl/debugfs.c
> index 2ff10a9..43a1a27 100644
> --- a/drivers/misc/cxl/debugfs.c
> +++ b/drivers/misc/cxl/debugfs.c
> @@ -94,6 +94,9 @@ void cxl_debugfs_adapter_remove(struct cxl *adapter)
>
> void cxl_debugfs_add_afu_regs_psl8(struct cxl_afu *afu, struct dentry *dir)
> {
> + debugfs_create_io_x64("sstp0", S_IRUSR, dir, _cxl_p2n_addr(afu, CXL_SSTP0_An));
> + debugfs_create_io_x64("sstp1", S_IRUSR, dir, _cxl_p2n_addr(afu, CXL_SSTP1_An));
> +
> debugfs_create_io_x64("fir", S_IRUSR, dir, _cxl_p1n_addr(afu, CXL_PSL_FIR_SLICE_An));
> debugfs_create_io_x64("serr", S_IRUSR, dir, _cxl_p1n_addr(afu, CXL_PSL_SERR_An));
> debugfs_create_io_x64("afu_debug", S_IRUSR, dir, _cxl_p1n_addr(afu, CXL_AFU_DEBUG_An));
> @@ -117,8 +120,7 @@ int cxl_debugfs_afu_add(struct cxl_afu *afu)
> debugfs_create_io_x64("sr", S_IRUSR, dir, _cxl_p1n_addr(afu, CXL_PSL_SR_An));
> debugfs_create_io_x64("dsisr", S_IRUSR, dir, _cxl_p2n_addr(afu, CXL_PSL_DSISR_An));
> debugfs_create_io_x64("dar", S_IRUSR, dir, _cxl_p2n_addr(afu, CXL_PSL_DAR_An));
> - debugfs_create_io_x64("sstp0", S_IRUSR, dir, _cxl_p2n_addr(afu, CXL_SSTP0_An));
> - debugfs_create_io_x64("sstp1", S_IRUSR, dir, _cxl_p2n_addr(afu, CXL_SSTP1_An));
> +
> debugfs_create_io_x64("err_status", S_IRUSR, dir, _cxl_p2n_addr(afu, CXL_PSL_ErrStat_An));
>
> if (afu->adapter->native->sl_ops->debugfs_add_afu_regs)
> diff --git a/drivers/misc/cxl/fault.c b/drivers/misc/cxl/fault.c
> index 14a5bfa..9b72b90 100644
> --- a/drivers/misc/cxl/fault.c
> +++ b/drivers/misc/cxl/fault.c
> @@ -231,12 +231,14 @@ void cxl_handle_fault(struct work_struct *fault_work)
> }
> }
>
> - if (dsisr & CXL_PSL_DSISR_An_DS)
> - cxl_handle_segment_miss(ctx, mm, dar);
> - else if (dsisr & CXL_PSL_DSISR_An_DM)
> - cxl_handle_page_fault(ctx, mm, dsisr, dar);
> - else
> - WARN(1, "cxl_handle_fault has nothing to handle\n");
> + if (cxl_is_psl8(ctx->afu)) {
> + if (dsisr & CXL_PSL_DSISR_An_DS)
> + cxl_handle_segment_miss(ctx, mm, dar);
> + else if (dsisr & CXL_PSL_DSISR_An_DM)
> + cxl_handle_page_fault(ctx, mm, dsisr, dar);
> + else
> + WARN(1, "cxl_handle_fault has nothing to handle\n");
> + }
>
> if (mm)
> mmput(mm);
> diff --git a/drivers/misc/cxl/native.c b/drivers/misc/cxl/native.c
> index ea08605..51aefcb 100644
> --- a/drivers/misc/cxl/native.c
> +++ b/drivers/misc/cxl/native.c
> @@ -155,16 +155,25 @@ int cxl_psl_purge(struct cxl_afu *afu)
> }
>
> dsisr = cxl_p2n_read(afu, CXL_PSL_DSISR_An);
> - pr_devel_ratelimited("PSL purging... PSL_CNTL: 0x%016llx PSL_DSISR: 0x%016llx\n", PSL_CNTL, dsisr);
> - if (dsisr & CXL_PSL_DSISR_TRANS) {
> - dar = cxl_p2n_read(afu, CXL_PSL_DAR_An);
> - dev_notice(&afu->dev, "PSL purge terminating pending translation, DSISR: 0x%016llx, DAR: 0x%016llx\n", dsisr, dar);
> - cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_AE);
> - } else if (dsisr) {
> - dev_notice(&afu->dev, "PSL purge acknowledging pending non-translation fault, DSISR: 0x%016llx\n", dsisr);
> - cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_A);
> - } else {
> - cpu_relax();
> + pr_devel_ratelimited("PSL purging... PSL_CNTL: 0x%016llx"
> + " PSL_DSISR: 0x%016llx\n",
> + PSL_CNTL, dsisr);
> + if (cxl_is_psl8(afu)) {
> + if (dsisr & CXL_PSL_DSISR_TRANS) {
> + dar = cxl_p2n_read(afu, CXL_PSL_DAR_An);
> + dev_notice(&afu->dev, "PSL purge terminating "
> + "pending translation, "
> + "DSISR: 0x%016llx, DAR: 0x%016llx\n",
> + dsisr, dar);
> + cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_AE);
> + } else if (dsisr) {
> + dev_notice(&afu->dev, "PSL purge acknowledging "
> + "pending non-translation fault, "
> + "DSISR: 0x%016llx\n", dsisr);
> + cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_A);
> + } else {
> + cpu_relax();
> + }
> }
> PSL_CNTL = cxl_p1n_read(afu, CXL_PSL_SCNTL_An);
> }
> @@ -466,7 +475,8 @@ static int remove_process_element(struct cxl_context *ctx)
>
> if (!rc)
> ctx->pe_inserted = false;
> - slb_invalid(ctx);
> + if (cxl_is_power8())
> + slb_invalid(ctx);
> pr_devel("%s Remove pe: %i finished\n", __func__, ctx->pe);
> mutex_unlock(&ctx->afu->native->spa_mutex);
>
> @@ -499,7 +509,8 @@ static int activate_afu_directed(struct cxl_afu *afu)
> attach_spa(afu);
>
> cxl_p1n_write(afu, CXL_PSL_SCNTL_An, CXL_PSL_SCNTL_An_PM_AFU);
> - cxl_p1n_write(afu, CXL_PSL_AMOR_An, 0xFFFFFFFFFFFFFFFFULL);
> + if (cxl_is_power8())
> + cxl_p1n_write(afu, CXL_PSL_AMOR_An, 0xFFFFFFFFFFFFFFFFULL);
> cxl_p1n_write(afu, CXL_PSL_ID_An, CXL_PSL_ID_An_F | CXL_PSL_ID_An_L);
>
> afu->current_mode = CXL_MODE_DIRECTED;
> @@ -872,7 +883,8 @@ static int native_get_irq_info(struct cxl_afu *afu, struct cxl_irq_info *info)
>
> info->dsisr = cxl_p2n_read(afu, CXL_PSL_DSISR_An);
> info->dar = cxl_p2n_read(afu, CXL_PSL_DAR_An);
> - info->dsr = cxl_p2n_read(afu, CXL_PSL_DSR_An);
> + if (cxl_is_power8())
> + info->dsr = cxl_p2n_read(afu, CXL_PSL_DSR_An);
> info->afu_err = cxl_p2n_read(afu, CXL_AFU_ERR_An);
> info->errstat = cxl_p2n_read(afu, CXL_PSL_ErrStat_An);
> info->proc_handle = 0;
> @@ -984,7 +996,8 @@ static void native_irq_wait(struct cxl_context *ctx)
> if (ph != ctx->pe)
> return;
> dsisr = cxl_p2n_read(ctx->afu, CXL_PSL_DSISR_An);
> - if ((dsisr & CXL_PSL_DSISR_PENDING) == 0)
> + if (cxl_is_psl8(ctx->afu) &&
> + ((dsisr & CXL_PSL_DSISR_PENDING) == 0))
> return;
> /*
> * We are waiting for the workqueue to process our
> @@ -1001,21 +1014,25 @@ static void native_irq_wait(struct cxl_context *ctx)
> static irqreturn_t native_slice_irq_err(int irq, void *data)
> {
> struct cxl_afu *afu = data;
> - u64 fir_slice, errstat, serr, afu_debug, afu_error, dsisr;
> + u64 errstat, serr, afu_error, dsisr;
> + u64 fir_slice, afu_debug;
>
> /*
> * slice err interrupt is only used with full PSL (no XSL)
> */
> serr = cxl_p1n_read(afu, CXL_PSL_SERR_An);
> - fir_slice = cxl_p1n_read(afu, CXL_PSL_FIR_SLICE_An);
> errstat = cxl_p2n_read(afu, CXL_PSL_ErrStat_An);
> - afu_debug = cxl_p1n_read(afu, CXL_AFU_DEBUG_An);
> afu_error = cxl_p2n_read(afu, CXL_AFU_ERR_An);
> dsisr = cxl_p2n_read(afu, CXL_PSL_DSISR_An);
> cxl_afu_decode_psl_serr(afu, serr);
> - dev_crit(&afu->dev, "PSL_FIR_SLICE_An: 0x%016llx\n", fir_slice);
> +
> + if (cxl_is_power8()) {
> + fir_slice = cxl_p1n_read(afu, CXL_PSL_FIR_SLICE_An);
> + afu_debug = cxl_p1n_read(afu, CXL_AFU_DEBUG_An);
> + dev_crit(&afu->dev, "PSL_FIR_SLICE_An: 0x%016llx\n", fir_slice);
> + dev_crit(&afu->dev, "CXL_PSL_AFU_DEBUG_An: 0x%016llx\n", afu_debug);
> + }
> dev_crit(&afu->dev, "CXL_PSL_ErrStat_An: 0x%016llx\n", errstat);
> - dev_crit(&afu->dev, "CXL_PSL_AFU_DEBUG_An: 0x%016llx\n", afu_debug);
> dev_crit(&afu->dev, "AFU_ERR_An: 0x%.16llx\n", afu_error);
> dev_crit(&afu->dev, "PSL_DSISR_An: 0x%.16llx\n", dsisr);
>
> @@ -1108,7 +1125,8 @@ int cxl_native_register_serr_irq(struct cxl_afu *afu)
> }
>
> serr = cxl_p1n_read(afu, CXL_PSL_SERR_An);
> - serr = (serr & 0x00ffffffffff0000ULL) | (afu->serr_hwirq & 0xffff);
> + if (cxl_is_power8())
> + serr = (serr & 0x00ffffffffff0000ULL) | (afu->serr_hwirq & 0xffff);
> cxl_p1n_write(afu, CXL_PSL_SERR_An, serr);
>
> return 0;
> diff --git a/drivers/misc/cxl/pci.c b/drivers/misc/cxl/pci.c
> index b952527..5880092 100644
> --- a/drivers/misc/cxl/pci.c
> +++ b/drivers/misc/cxl/pci.c
> @@ -324,32 +324,33 @@ static void dump_afu_descriptor(struct cxl_afu *afu)
> #undef show_reg
> }
>
> -#define CAPP_UNIT0_ID 0xBA
> -#define CAPP_UNIT1_ID 0XBE
> +#define P8_CAPP_UNIT0_ID 0xBA
> +#define P8_CAPP_UNIT1_ID 0XBE
>
> static u64 get_capp_unit_id(struct device_node *np)
> {
> u32 phb_index;
>
> - /*
> - * For chips other than POWER8NVL, we only have CAPP 0,
> - * irrespective of which PHB is used.
> - */
> - if (!pvr_version_is(PVR_POWER8NVL))
> - return CAPP_UNIT0_ID;
> + if (of_property_read_u32(np, "ibm,phb-index", &phb_index))
> + return 0;
>
> /*
> - * For POWER8NVL, assume CAPP 0 is attached to PHB0 and
> - * CAPP 1 is attached to PHB1.
> + * POWER 8:
> + * - For chips other than POWER8NVL, we only have CAPP 0,
> + * irrespective of which PHB is used.
> + * - For POWER8NVL, assume CAPP 0 is attached to PHB0 and
> + * CAPP 1 is attached to PHB1.
> */
> - if (of_property_read_u32(np, "ibm,phb-index", &phb_index))
> - return 0;
> + if (cxl_is_power8()) {
> + if (!pvr_version_is(PVR_POWER8NVL))
> + return P8_CAPP_UNIT0_ID;
>
> - if (phb_index == 0)
> - return CAPP_UNIT0_ID;
> + if (phb_index == 0)
> + return P8_CAPP_UNIT0_ID;
>
> - if (phb_index == 1)
> - return CAPP_UNIT1_ID;
> + if (phb_index == 1)
> + return P8_CAPP_UNIT1_ID;
> + }
>
> return 0;
> }
> @@ -968,7 +969,7 @@ static int cxl_afu_descriptor_looks_ok(struct cxl_afu *afu)
> }
>
> if (afu->pp_psa && (afu->pp_size < PAGE_SIZE))
> - dev_warn(&afu->dev, "AFU uses < PAGE_SIZE per-process PSA!");
> + dev_warn(&afu->dev, "AFU uses pp_size(%#016llx) < PAGE_SIZE per-process PSA!\n", afu->pp_size);
>
> for (i = 0; i < afu->crs_num; i++) {
> rc = cxl_ops->afu_cr_read32(afu, i, 0, &val);
> @@ -1251,8 +1252,13 @@ int cxl_pci_reset(struct cxl *adapter)
>
> dev_info(&dev->dev, "CXL reset\n");
>
> - /* the adapter is about to be reset, so ignore errors */
> - cxl_data_cache_flush(adapter);
> + /*
> + * The adapter is about to be reset, so ignore errors.
> + * Not supported on P9 DD1 but don't forget to enable it
> + * on P9 DD2
> + */
> + if (cxl_is_power8())
> + cxl_data_cache_flush(adapter);
>
> /* pcie_warm_reset requests a fundamental pci reset which includes a
> * PERST assert/deassert. PERST triggers a loading of the image
> @@ -1382,6 +1388,14 @@ static void cxl_fixup_malformed_tlp(struct cxl *adapter, struct pci_dev *dev)
> pci_write_config_dword(dev, aer + PCI_ERR_UNCOR_MASK, data);
> }
>
> +static bool cxl_compatible_caia_version(struct cxl *adapter)
> +{
> + if (cxl_is_power8() && (adapter->caia_major == 1))
> + return true;
> +
> + return false;
> +}
> +
> static int cxl_vsec_looks_ok(struct cxl *adapter, struct pci_dev *dev)
> {
> if (adapter->vsec_status & CXL_STATUS_SECOND_PORT)
> @@ -1392,6 +1406,12 @@ static int cxl_vsec_looks_ok(struct cxl *adapter, struct pci_dev *dev)
> return -EINVAL;
> }
>
> + if (!cxl_compatible_caia_version(adapter)) {
> + dev_info(&dev->dev, "Ignoring card. PSL type is not supported "
> + "(caia version: %d)\n", adapter->caia_major);
> + return -ENODEV;
> + }
> +
> if (!adapter->slices) {
> /* Once we support dynamic reprogramming we can use the card if
> * it supports loadable AFUs */
> @@ -1566,8 +1586,10 @@ static void set_sl_ops(struct cxl *adapter, struct pci_dev *dev)
> adapter->native->sl_ops = &xsl_ops;
> adapter->min_pe = 1; /* Workaround for CX-4 hardware bug */
> } else {
> - dev_info(&dev->dev, "Device uses a PSL8\n");
> - adapter->native->sl_ops = &psl8_ops;
> + if (cxl_is_power8()) {
> + dev_info(&dev->dev, "Device uses a PSL8\n");
> + adapter->native->sl_ops = &psl8_ops;
> + }
> }
> }
>
^ permalink raw reply [flat|nested] 20+ messages in thread
* [PATCH V2 6/6] cxl: Add psl9 specific code
2017-03-14 11:08 [PATCH V2 0/6] cxl: Add support for Coherent Accelerator Interface Architecture 2.0 Christophe Lombard
` (4 preceding siblings ...)
2017-03-14 11:08 ` [PATCH V2 5/6] cxl: Isolate few psl8 specific calls Christophe Lombard
@ 2017-03-14 11:08 ` Christophe Lombard
2017-03-21 11:21 ` Frederic Barrat
2017-03-21 2:47 ` [PATCH V2 0/6] cxl: Add support for Coherent Accelerator Interface Architecture 2.0 Andrew Donnellan
6 siblings, 1 reply; 20+ messages in thread
From: Christophe Lombard @ 2017-03-14 11:08 UTC (permalink / raw)
To: linuxppc-dev, fbarrat, imunsie, andrew.donnellan
The new Coherent Accelerator Interface Architecture, level 2, for the
IBM POWER9 brings new content and features:
- POWER9 Service Layer
- Registers
- Radix mode
- Process element entry
- Dedicated-Shared Process Programming Model
- Translation Fault Handling
- CAPP
- Memory Context ID
If a valid mm_struct is found the memory context id is used for each
transaction associated with the process handle. The PSL uses the
context ID to find the corresponding process element.
Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com>
---
drivers/misc/cxl/context.c | 13 +++
drivers/misc/cxl/cxl.h | 124 ++++++++++++++++++++----
drivers/misc/cxl/debugfs.c | 19 ++++
drivers/misc/cxl/fault.c | 48 ++++++----
drivers/misc/cxl/guest.c | 8 +-
drivers/misc/cxl/irq.c | 53 +++++++++++
drivers/misc/cxl/native.c | 218 +++++++++++++++++++++++++++++++++++++++++--
drivers/misc/cxl/pci.c | 228 +++++++++++++++++++++++++++++++++++++++++++--
drivers/misc/cxl/trace.h | 43 +++++++++
9 files changed, 700 insertions(+), 54 deletions(-)
diff --git a/drivers/misc/cxl/context.c b/drivers/misc/cxl/context.c
index 0fd3cf8..8d41fc5 100644
--- a/drivers/misc/cxl/context.c
+++ b/drivers/misc/cxl/context.c
@@ -205,6 +205,19 @@ int cxl_context_iomap(struct cxl_context *ctx, struct vm_area_struct *vma)
return -EBUSY;
}
+ if ((ctx->afu->current_mode == CXL_MODE_DEDICATED) &&
+ (cxl_is_psl9(ctx->afu))) {
+ /* make sure there is a valid problem state area space for this AFU */
+ if (ctx->master && !ctx->afu->psa) {
+ pr_devel("AFU doesn't support mmio space\n");
+ return -EINVAL;
+ }
+
+ /* Can't mmap until the AFU is enabled */
+ if (!ctx->afu->enabled)
+ return -EBUSY;
+ }
+
pr_devel("%s: mmio physical: %llx pe: %i master:%i\n", __func__,
ctx->psn_phys, ctx->pe , ctx->master);
diff --git a/drivers/misc/cxl/cxl.h b/drivers/misc/cxl/cxl.h
index f6a3a34..fbdc511 100644
--- a/drivers/misc/cxl/cxl.h
+++ b/drivers/misc/cxl/cxl.h
@@ -63,7 +63,7 @@ typedef struct {
/* Memory maps. Ref CXL Appendix A */
/* PSL Privilege 1 Memory Map */
-/* Configuration and Control area */
+/* Configuration and Control area - CAIA 1&2 */
static const cxl_p1_reg_t CXL_PSL_CtxTime = {0x0000};
static const cxl_p1_reg_t CXL_PSL_ErrIVTE = {0x0008};
static const cxl_p1_reg_t CXL_PSL_KEY1 = {0x0010};
@@ -98,11 +98,29 @@ static const cxl_p1_reg_t CXL_XSL_Timebase = {0x0100};
static const cxl_p1_reg_t CXL_XSL_TB_CTLSTAT = {0x0108};
static const cxl_p1_reg_t CXL_XSL_FEC = {0x0158};
static const cxl_p1_reg_t CXL_XSL_DSNCTL = {0x0168};
+/* PSL registers - CAIA 2 */
+static const cxl_p1_reg_t CXL_PSL9_CONTROL = {0x0020};
+static const cxl_p1_reg_t CXL_XSL9_DSNCTL = {0x0168};
+static const cxl_p1_reg_t CXL_PSL9_FIR1 = {0x0300};
+static const cxl_p1_reg_t CXL_PSL9_FIR2 = {0x0308}; /* TBD NML CAIA 2 */
+static const cxl_p1_reg_t CXL_PSL9_Timebase = {0x0310};
+static const cxl_p1_reg_t CXL_PSL9_DEBUG = {0x0320};
+static const cxl_p1_reg_t CXL_PSL9_FIR_CNTL = {0x0348};
+static const cxl_p1_reg_t CXL_PSL9_DSNDCTL = {0x0350};
+static const cxl_p1_reg_t CXL_PSL9_TB_CTLSTAT = {0x0340};
+static const cxl_p1_reg_t CXL_PSL9_TRACECFG = {0x0368};
+static const cxl_p1_reg_t CXL_PSL9_APCDEDALLOC = {0x0378};
+static const cxl_p1_reg_t CXL_PSL9_APCDEDTYPE = {0x0380};
+static const cxl_p1_reg_t CXL_PSL9_TNR_ADDR = {0x0388};
+static const cxl_p1_reg_t CXL_PSL9_GP_CT = {0x0398};
+static const cxl_p1_reg_t CXL_XSL9_IERAT = {0x0588};
+static const cxl_p1_reg_t CXL_XSL9_ILPP = {0x0590};
+
/* 0x7F00:7FFF Reserved PCIe MSI-X Pending Bit Array area */
/* 0x8000:FFFF Reserved PCIe MSI-X Table Area */
/* PSL Slice Privilege 1 Memory Map */
-/* Configuration Area */
+/* Configuration Area - CAIA 1&2 */
static const cxl_p1n_reg_t CXL_PSL_SR_An = {0x00};
static const cxl_p1n_reg_t CXL_PSL_LPID_An = {0x08};
static const cxl_p1n_reg_t CXL_PSL_AMBAR_An = {0x10};
@@ -111,17 +129,18 @@ static const cxl_p1n_reg_t CXL_PSL_ID_An = {0x20};
static const cxl_p1n_reg_t CXL_PSL_SERR_An = {0x28};
/* Memory Management and Lookaside Buffer Management - CAIA 1*/
static const cxl_p1n_reg_t CXL_PSL_SDR_An = {0x30};
+/* Memory Management and Lookaside Buffer Management - CAIA 1&2 */
static const cxl_p1n_reg_t CXL_PSL_AMOR_An = {0x38};
-/* Pointer Area */
+/* Pointer Area - CAIA 1&2 */
static const cxl_p1n_reg_t CXL_HAURP_An = {0x80};
static const cxl_p1n_reg_t CXL_PSL_SPAP_An = {0x88};
static const cxl_p1n_reg_t CXL_PSL_LLCMD_An = {0x90};
-/* Control Area */
+/* Control Area - CAIA 1&2 */
static const cxl_p1n_reg_t CXL_PSL_SCNTL_An = {0xA0};
static const cxl_p1n_reg_t CXL_PSL_CtxTime_An = {0xA8};
static const cxl_p1n_reg_t CXL_PSL_IVTE_Offset_An = {0xB0};
static const cxl_p1n_reg_t CXL_PSL_IVTE_Limit_An = {0xB8};
-/* 0xC0:FF Implementation Dependent Area */
+/* 0xC0:FF Implementation Dependent Area - CAIA 1&2 */
static const cxl_p1n_reg_t CXL_PSL_FIR_SLICE_An = {0xC0};
static const cxl_p1n_reg_t CXL_AFU_DEBUG_An = {0xC8};
/* 0xC0:FF Implementation Dependent Area - CAIA 1 */
@@ -131,7 +150,7 @@ static const cxl_p1n_reg_t CXL_PSL_RXCTL_A = {0xE0};
static const cxl_p1n_reg_t CXL_PSL_SLICE_TRACE = {0xE8};
/* PSL Slice Privilege 2 Memory Map */
-/* Configuration and Control Area */
+/* Configuration and Control Area - CAIA 1&2 */
static const cxl_p2n_reg_t CXL_PSL_PID_TID_An = {0x000};
static const cxl_p2n_reg_t CXL_CSRP_An = {0x008};
/* Configuration and Control Area - CAIA 1 */
@@ -145,17 +164,17 @@ static const cxl_p2n_reg_t CXL_PSL_AMR_An = {0x030};
static const cxl_p2n_reg_t CXL_SLBIE_An = {0x040};
static const cxl_p2n_reg_t CXL_SLBIA_An = {0x048};
static const cxl_p2n_reg_t CXL_SLBI_Select_An = {0x050};
-/* Interrupt Registers */
+/* Interrupt Registers - CAIA 1&2 */
static const cxl_p2n_reg_t CXL_PSL_DSISR_An = {0x060};
static const cxl_p2n_reg_t CXL_PSL_DAR_An = {0x068};
static const cxl_p2n_reg_t CXL_PSL_DSR_An = {0x070};
static const cxl_p2n_reg_t CXL_PSL_TFC_An = {0x078};
static const cxl_p2n_reg_t CXL_PSL_PEHandle_An = {0x080};
static const cxl_p2n_reg_t CXL_PSL_ErrStat_An = {0x088};
-/* AFU Registers */
+/* AFU Registers - CAIA 1&2 */
static const cxl_p2n_reg_t CXL_AFU_Cntl_An = {0x090};
static const cxl_p2n_reg_t CXL_AFU_ERR_An = {0x098};
-/* Work Element Descriptor */
+/* Work Element Descriptor - CAIA 1&2 */
static const cxl_p2n_reg_t CXL_PSL_WED_An = {0x0A0};
/* 0x0C0:FFF Implementation Dependent Area */
@@ -182,6 +201,10 @@ static const cxl_p2n_reg_t CXL_PSL_WED_An = {0x0A0};
#define CXL_PSL_SR_An_SF MSR_SF /* 64bit */
#define CXL_PSL_SR_An_TA (1ull << (63-1)) /* Tags active, GA1: 0 */
#define CXL_PSL_SR_An_HV MSR_HV /* Hypervisor, GA1: 0 */
+#define CXL_PSL_SR_An_XLAT_hpt (0ull << (63-6))/* Hashed page table (HPT) mode */
+#define CXL_PSL_SR_An_XLAT_roh (2ull << (63-6))/* Radix on HPT mode */
+#define CXL_PSL_SR_An_XLAT_ror (3ull << (63-6))/* Radix on Radix mode */
+#define CXL_PSL_SR_An_BOT (1ull << (63-10)) /* Use the in-memory segment table */
#define CXL_PSL_SR_An_PR MSR_PR /* Problem state, GA1: 1 */
#define CXL_PSL_SR_An_ISL (1ull << (63-53)) /* Ignore Segment Large Page */
#define CXL_PSL_SR_An_TC (1ull << (63-54)) /* Page Table secondary hash */
@@ -298,12 +321,38 @@ static const cxl_p2n_reg_t CXL_PSL_WED_An = {0x0A0};
#define CXL_PSL_DSISR_An_S DSISR_ISSTORE /* Access was afu_wr or afu_zero */
#define CXL_PSL_DSISR_An_K DSISR_KEYFAULT /* Access not permitted by virtual page class key protection */
+/****** CXL_PSL_DSISR_An - CAIA 2 ****************************************************/
+#define CXL_PSL9_DSISR_An_TF (1ull << (63-3)) /* Translation fault */
+#define CXL_PSL9_DSISR_An_PE (1ull << (63-4)) /* PSL Error (implementation specific) */
+#define CXL_PSL9_DSISR_An_AE (1ull << (63-5)) /* AFU Error */
+#define CXL_PSL9_DSISR_An_OC (1ull << (63-6)) /* OS Context Warning */
+#define CXL_PSL9_DSISR_An_S (1ull << (63-38)) /* TF for a write operation */
+#define CXL_PSL9_DSISR_PENDING (CXL_PSL9_DSISR_An_TF | CXL_PSL9_DSISR_An_PE | CXL_PSL9_DSISR_An_AE | CXL_PSL9_DSISR_An_OC)
+/* NOTE: Bits 56:63 (Checkout Response Status) are valid when DSISR_An[TF] = 1
+ * Status (0:7) Encoding
+ */
+#define CXL_PSL9_DSISR_An_CO_MASK 0x00000000000000ffULL
+#define CXL_PSL9_DSISR_An_SF 0x0000000000000080ULL /* Segment Fault 0b10000000 */
+#define CXL_PSL9_DSISR_An_PF_SLR 0x0000000000000088ULL /* PTE not found (Single Level Radix) 0b10001000 */
+#define CXL_PSL9_DSISR_An_PF_RGC 0x000000000000008CULL /* PTE not found (Radix Guest (child)) 0b10001100 */
+#define CXL_PSL9_DSISR_An_PF_RGP 0x0000000000000090ULL /* PTE not found (Radix Guest (parent)) 0b10010000 */
+#define CXL_PSL9_DSISR_An_PF_HRH 0x0000000000000094ULL /* PTE not found (HPT/Radix Host) 0b10010100 */
+#define CXL_PSL9_DSISR_An_PF_STEG 0x000000000000009CULL /* PTE not found (STEG VA) 0b10011100 */
+
/****** CXL_PSL_TFC_An ******************************************************/
#define CXL_PSL_TFC_An_A (1ull << (63-28)) /* Acknowledge non-translation fault */
#define CXL_PSL_TFC_An_C (1ull << (63-29)) /* Continue (abort transaction) */
#define CXL_PSL_TFC_An_AE (1ull << (63-30)) /* Restart PSL with address error */
#define CXL_PSL_TFC_An_R (1ull << (63-31)) /* Restart PSL transaction */
+/****** CXL_XSL9_IERAT_ERAT - CAIA 2 **********************************/
+#define CXL_XSL9_IERAT_MLPID (1ull << (63-0)) /* Match LPID */
+#define CXL_XSL9_IERAT_MPID (1ull << (63-1)) /* Match PID */
+#define CXL_XSL9_IERAT_PRS (1ull << (63-4)) /* PRS bit for Radix invalidations */
+#define CXL_XSL9_IERAT_INVR (1ull << (63-3)) /* Invalidate Radix */
+#define CXL_XSL9_IERAT_IALL (1ull << (63-8)) /* Invalidate All */
+#define CXL_XSL9_IERAT_IINPROG (1ull << (63-63)) /* Invalidate in progress */
+
/* cxl_process_element->software_status */
#define CXL_PE_SOFTWARE_STATE_V (1ul << (31 - 0)) /* Valid */
#define CXL_PE_SOFTWARE_STATE_C (1ul << (31 - 29)) /* Complete */
@@ -654,25 +703,38 @@ int cxl_pci_reset(struct cxl *adapter);
void cxl_pci_release_afu(struct device *dev);
ssize_t cxl_pci_read_adapter_vpd(struct cxl *adapter, void *buf, size_t len);
-/* common == phyp + powernv */
+/* common == phyp + powernv - CAIA 1&2 */
struct cxl_process_element_common {
__be32 tid;
__be32 pid;
__be64 csrp;
- __be64 aurp0;
- __be64 aurp1;
- __be64 sstp0;
- __be64 sstp1;
+ union {
+ struct {
+ __be64 aurp0;
+ __be64 aurp1;
+ __be64 sstp0;
+ __be64 sstp1;
+ } psl8; /* CAIA 1 */
+ struct {
+ u8 reserved2[8];
+ u8 reserved3[8];
+ u8 reserved4[8];
+ u8 reserved5[8];
+ } psl9; /* CAIA 2 */
+ } u;
__be64 amr;
- u8 reserved3[4];
+ u8 reserved6[4];
__be64 wed;
} __packed;
-/* just powernv */
+/* just powernv - CAIA 1&2 */
struct cxl_process_element {
__be64 sr;
__be64 SPOffset;
- __be64 sdr;
+ union {
+ __be64 sdr; /* CAIA 1 */
+ u8 reserved1[8]; /* CAIA 2 */
+ } u;
__be64 haurp;
__be32 ctxtime;
__be16 ivte_offsets[4];
@@ -761,6 +823,16 @@ static inline bool cxl_is_power8(void)
return false;
}
+static inline bool cxl_is_power9(void)
+{
+ /* intermediate solution */
+ if (!cxl_is_power8() &&
+ (cpu_has_feature(CPU_FTRS_POWER9) ||
+ cpu_has_feature(CPU_FTR_POWER9_DD1)))
+ return true;
+ return false;
+}
+
static inline bool cxl_is_psl8(struct cxl_afu *afu)
{
if (afu->adapter->caia_major == 1)
@@ -768,6 +840,13 @@ static inline bool cxl_is_psl8(struct cxl_afu *afu)
return false;
}
+static inline bool cxl_is_psl9(struct cxl_afu *afu)
+{
+ if (afu->adapter->caia_major == 2)
+ return true;
+ return false;
+}
+
ssize_t cxl_pci_afu_read_err_buffer(struct cxl_afu *afu, char *buf,
loff_t off, size_t count);
@@ -834,9 +913,13 @@ void afu_irq_name_free(struct cxl_context *ctx);
#ifdef CONFIG_DEBUG_FS
+int cxl_attach_afu_directed_psl9(struct cxl_context *ctx, u64 wed, u64 amr);
int cxl_attach_afu_directed_psl8(struct cxl_context *ctx, u64 wed, u64 amr);
+int cxl_activate_dedicated_process_psl9(struct cxl_afu *afu);
int cxl_activate_dedicated_process_psl8(struct cxl_afu *afu);
+int cxl_attach_dedicated_process_psl9(struct cxl_context *ctx, u64 wed, u64 amr);
int cxl_attach_dedicated_process_psl8(struct cxl_context *ctx, u64 wed, u64 amr);
+void cxl_update_dedicated_ivtes_psl9(struct cxl_context *ctx);
void cxl_update_dedicated_ivtes_psl8(struct cxl_context *ctx);
int cxl_debugfs_init(void);
@@ -845,9 +928,12 @@ int cxl_debugfs_adapter_add(struct cxl *adapter);
void cxl_debugfs_adapter_remove(struct cxl *adapter);
int cxl_debugfs_afu_add(struct cxl_afu *afu);
void cxl_debugfs_afu_remove(struct cxl_afu *afu);
+void cxl_stop_trace_psl9(struct cxl *cxl);
void cxl_stop_trace_psl8(struct cxl *cxl);
+void cxl_debugfs_add_adapter_regs_psl9(struct cxl *adapter, struct dentry *dir);
void cxl_debugfs_add_adapter_regs_psl8(struct cxl *adapter, struct dentry *dir);
void cxl_debugfs_add_adapter_regs_xsl(struct cxl *adapter, struct dentry *dir);
+void cxl_debugfs_add_afu_regs_psl9(struct cxl_afu *afu, struct dentry *dir);
void cxl_debugfs_add_afu_regs_psl8(struct cxl_afu *afu, struct dentry *dir);
#else /* CONFIG_DEBUG_FS */
@@ -950,8 +1036,11 @@ struct cxl_irq_info {
};
void cxl_assign_psn_space(struct cxl_context *ctx);
+int cxl_invalidate_all_psl9(struct cxl *adapter);
int cxl_invalidate_all_psl8(struct cxl *adapter);
+irqreturn_t cxl_irq_psl9(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info);
irqreturn_t cxl_irq_psl8(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info);
+irqreturn_t cxl_fail_irq_psl9(struct cxl_afu *afu, struct cxl_irq_info *irq_info);
irqreturn_t cxl_fail_irq_psl8(struct cxl_afu *afu, struct cxl_irq_info *irq_info);
int cxl_register_one_irq(struct cxl *adapter, irq_handler_t handler,
void *cookie, irq_hw_number_t *dest_hwirq,
@@ -963,6 +1052,7 @@ int cxl_data_cache_flush(struct cxl *adapter);
int cxl_afu_disable(struct cxl_afu *afu);
int cxl_psl_purge(struct cxl_afu *afu);
+void cxl_native_irq_dump_regs_psl9(struct cxl_context *ctx);
void cxl_native_irq_dump_regs_psl8(struct cxl_context *ctx);
void cxl_native_err_irq_dump_regs(struct cxl *adapter);
int cxl_pci_vphb_add(struct cxl_afu *afu);
diff --git a/drivers/misc/cxl/debugfs.c b/drivers/misc/cxl/debugfs.c
index 43a1a27..eae9d74 100644
--- a/drivers/misc/cxl/debugfs.c
+++ b/drivers/misc/cxl/debugfs.c
@@ -15,6 +15,12 @@
static struct dentry *cxl_debugfs;
+void cxl_stop_trace_psl9(struct cxl *adapter)
+{
+ /* Stop the trace */
+ cxl_p1_write(adapter, CXL_PSL9_TRACECFG, 0x4480000000000000ULL);
+}
+
void cxl_stop_trace_psl8(struct cxl *adapter)
{
int slice;
@@ -53,6 +59,14 @@ static struct dentry *debugfs_create_io_x64(const char *name, umode_t mode,
(void __force *)value, &fops_io_x64);
}
+void cxl_debugfs_add_adapter_regs_psl9(struct cxl *adapter, struct dentry *dir)
+{
+ debugfs_create_io_x64("fir1", S_IRUSR, dir, _cxl_p1_addr(adapter, CXL_PSL9_FIR1));
+ debugfs_create_io_x64("fir2", S_IRUSR, dir, _cxl_p1_addr(adapter, CXL_PSL9_FIR2));
+ debugfs_create_io_x64("fir_cntl", S_IRUSR, dir, _cxl_p1_addr(adapter, CXL_PSL9_FIR_CNTL));
+ debugfs_create_io_x64("trace", S_IRUSR | S_IWUSR, dir, _cxl_p1_addr(adapter, CXL_PSL9_TRACECFG));
+}
+
void cxl_debugfs_add_adapter_regs_psl8(struct cxl *adapter, struct dentry *dir)
{
debugfs_create_io_x64("fir1", S_IRUSR, dir, _cxl_p1_addr(adapter, CXL_PSL_FIR1));
@@ -92,6 +106,11 @@ void cxl_debugfs_adapter_remove(struct cxl *adapter)
debugfs_remove_recursive(adapter->debugfs);
}
+void cxl_debugfs_add_afu_regs_psl9(struct cxl_afu *afu, struct dentry *dir)
+{
+ debugfs_create_io_x64("serr", S_IRUSR, dir, _cxl_p1n_addr(afu, CXL_PSL_SERR_An));
+}
+
void cxl_debugfs_add_afu_regs_psl8(struct cxl_afu *afu, struct dentry *dir)
{
debugfs_create_io_x64("sstp0", S_IRUSR, dir, _cxl_p2n_addr(afu, CXL_SSTP0_An));
diff --git a/drivers/misc/cxl/fault.c b/drivers/misc/cxl/fault.c
index 9b72b90..3f84960 100644
--- a/drivers/misc/cxl/fault.c
+++ b/drivers/misc/cxl/fault.c
@@ -146,25 +146,26 @@ static void cxl_handle_page_fault(struct cxl_context *ctx,
return cxl_ack_ae(ctx);
}
- /*
- * update_mmu_cache() will not have loaded the hash since current->trap
- * is not a 0x400 or 0x300, so just call hash_page_mm() here.
- */
- access = _PAGE_PRESENT | _PAGE_READ;
- if (dsisr & CXL_PSL_DSISR_An_S)
- access |= _PAGE_WRITE;
-
- access |= _PAGE_PRIVILEGED;
- if ((!ctx->kernel) || (REGION_ID(dar) == USER_REGION_ID))
- access &= ~_PAGE_PRIVILEGED;
-
- if (dsisr & DSISR_NOHPTE)
- inv_flags |= HPTE_NOHPTE_UPDATE;
-
- local_irq_save(flags);
- hash_page_mm(mm, dar, access, 0x300, inv_flags);
- local_irq_restore(flags);
-
+ if (!radix_enabled()) {
+ /*
+ * update_mmu_cache() will not have loaded the hash since current->trap
+ * is not a 0x400 or 0x300, so just call hash_page_mm() here.
+ */
+ access = _PAGE_PRESENT | _PAGE_READ;
+ if (dsisr & CXL_PSL_DSISR_An_S)
+ access |= _PAGE_WRITE;
+
+ access |= _PAGE_PRIVILEGED;
+ if ((!ctx->kernel) || (REGION_ID(dar) == USER_REGION_ID))
+ access &= ~_PAGE_PRIVILEGED;
+
+ if (dsisr & DSISR_NOHPTE)
+ inv_flags |= HPTE_NOHPTE_UPDATE;
+
+ local_irq_save(flags);
+ hash_page_mm(mm, dar, access, 0x300, inv_flags);
+ local_irq_restore(flags);
+ }
pr_devel("Page fault successfully handled for pe: %i!\n", ctx->pe);
cxl_ops->ack_irq(ctx, CXL_PSL_TFC_An_R, 0);
}
@@ -239,6 +240,15 @@ void cxl_handle_fault(struct work_struct *fault_work)
else
WARN(1, "cxl_handle_fault has nothing to handle\n");
}
+ if (cxl_is_psl9(ctx->afu)) {
+ if ((dsisr & CXL_PSL9_DSISR_An_CO_MASK) &
+ (CXL_PSL9_DSISR_An_PF_SLR | CXL_PSL9_DSISR_An_PF_RGC |
+ CXL_PSL9_DSISR_An_PF_RGP | CXL_PSL9_DSISR_An_PF_HRH |
+ CXL_PSL9_DSISR_An_PF_STEG))
+ cxl_handle_page_fault(ctx, mm, dsisr, dar);
+ else
+ WARN(1, "cxl_handle_fault has nothing to handle\n");
+ }
if (mm)
mmput(mm);
diff --git a/drivers/misc/cxl/guest.c b/drivers/misc/cxl/guest.c
index 3ad7381..f58b4b6c 100644
--- a/drivers/misc/cxl/guest.c
+++ b/drivers/misc/cxl/guest.c
@@ -551,13 +551,13 @@ static int attach_afu_directed(struct cxl_context *ctx, u64 wed, u64 amr)
elem->common.tid = cpu_to_be32(0); /* Unused */
elem->common.pid = cpu_to_be32(pid);
elem->common.csrp = cpu_to_be64(0); /* disable */
- elem->common.aurp0 = cpu_to_be64(0); /* disable */
- elem->common.aurp1 = cpu_to_be64(0); /* disable */
+ elem->common.u.psl8.aurp0 = cpu_to_be64(0); /* disable */
+ elem->common.u.psl8.aurp1 = cpu_to_be64(0); /* disable */
cxl_prefault(ctx, wed);
- elem->common.sstp0 = cpu_to_be64(ctx->sstp0);
- elem->common.sstp1 = cpu_to_be64(ctx->sstp1);
+ elem->common.u.psl8.sstp0 = cpu_to_be64(ctx->sstp0);
+ elem->common.u.psl8.sstp1 = cpu_to_be64(ctx->sstp1);
/*
* Ensure we have at least one interrupt allocated to take faults for
diff --git a/drivers/misc/cxl/irq.c b/drivers/misc/cxl/irq.c
index fa9f8a2..1eb5168 100644
--- a/drivers/misc/cxl/irq.c
+++ b/drivers/misc/cxl/irq.c
@@ -34,6 +34,59 @@ static irqreturn_t schedule_cxl_fault(struct cxl_context *ctx, u64 dsisr, u64 da
return IRQ_HANDLED;
}
+irqreturn_t cxl_irq_psl9(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info)
+{
+ u64 dsisr, dar;
+
+ dsisr = irq_info->dsisr;
+ dar = irq_info->dar;
+
+ trace_cxl_psl9_irq(ctx, irq, dsisr, dar);
+
+ pr_devel("CXL interrupt %i for afu pe: %i DSISR: %#llx DAR: %#llx\n", irq, ctx->pe, dsisr, dar);
+
+ if (dsisr & CXL_PSL9_DSISR_An_TF) {
+ pr_devel("CXL interrupt: Scheduling translation fault"
+ " handling for later (pe: %i)\n", ctx->pe);
+ return schedule_cxl_fault(ctx, dsisr, dar);
+ }
+
+ if (dsisr & CXL_PSL9_DSISR_An_PE)
+ return cxl_ops->handle_psl_slice_error(ctx, dsisr,
+ irq_info->errstat);
+ if (dsisr & CXL_PSL9_DSISR_An_AE) {
+ pr_devel("CXL interrupt: AFU Error 0x%016llx\n", irq_info->afu_err);
+
+ if (ctx->pending_afu_err) {
+ /*
+ * This shouldn't happen - the PSL treats these errors
+ * as fatal and will have reset the AFU, so there's not
+ * much point buffering multiple AFU errors.
+ * OTOH if we DO ever see a storm of these come in it's
+ * probably best that we log them somewhere:
+ */
+ dev_err_ratelimited(&ctx->afu->dev, "CXL AFU Error "
+ "undelivered to pe %i: 0x%016llx\n",
+ ctx->pe, irq_info->afu_err);
+ } else {
+ spin_lock(&ctx->lock);
+ ctx->afu_err = irq_info->afu_err;
+ ctx->pending_afu_err = 1;
+ spin_unlock(&ctx->lock);
+
+ wake_up_all(&ctx->wq);
+ }
+
+ cxl_ops->ack_irq(ctx, CXL_PSL_TFC_An_A, 0);
+ return IRQ_HANDLED;
+ }
+ if (dsisr & CXL_PSL9_DSISR_An_OC)
+ pr_devel("CXL interrupt: OS Context Warning\n");
+
+ WARN(1, "Unhandled CXL PSL IRQ\n");
+ return IRQ_HANDLED;
+}
+
irqreturn_t cxl_irq_psl8(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info)
{
u64 dsisr, dar;
diff --git a/drivers/misc/cxl/native.c b/drivers/misc/cxl/native.c
index 51aefcb..b4f7569 100644
--- a/drivers/misc/cxl/native.c
+++ b/drivers/misc/cxl/native.c
@@ -175,6 +175,23 @@ int cxl_psl_purge(struct cxl_afu *afu)
cpu_relax();
}
}
+ if (cxl_is_psl9(afu)) {
+ if (dsisr & CXL_PSL9_DSISR_An_TF) {
+ dar = cxl_p2n_read(afu, CXL_PSL_DAR_An);
+ dev_notice(&afu->dev, "PSL purge terminating "
+ "pending translation, "
+ "DSISR: 0x%016llx, DAR: 0x%016llx\n",
+ dsisr, dar);
+ cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_AE);
+ } else if (dsisr) {
+ dev_notice(&afu->dev, "PSL purge acknowledging "
+ "pending non-translation fault, "
+ "DSISR: 0x%016llx\n", dsisr);
+ cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_A);
+ } else {
+ cpu_relax();
+ }
+ }
PSL_CNTL = cxl_p1n_read(afu, CXL_PSL_SCNTL_An);
}
end = local_clock();
@@ -267,6 +284,36 @@ void cxl_release_spa(struct cxl_afu *afu)
}
}
+int cxl_invalidate_all_psl9(struct cxl *adapter)
+{
+ unsigned long timeout = jiffies + (HZ * CXL_TIMEOUT);
+ u64 ierat;
+
+ /* do not invalidate ERAT entries when not reloading on PERST */
+ if (adapter->perst_loads_image)
+ return 0;
+
+ pr_devel("CXL adapter - invalidation of all ERAT entries\n");
+
+ /* Invalidates all ERAT entries for Radix or HPT */
+ ierat = CXL_XSL9_IERAT_IALL;
+ if (radix_enabled())
+ ierat |= CXL_XSL9_IERAT_INVR;
+ cxl_p1_write(adapter, CXL_XSL9_IERAT, ierat);
+
+ while (cxl_p1_read(adapter, CXL_XSL9_IERAT) & CXL_XSL9_IERAT_IINPROG) {
+ if (time_after_eq(jiffies, timeout)) {
+ dev_warn(&adapter->dev,
+ "WARNING: CXL adapter invalidation of all ERAT entries timed out!\n");
+ return -EBUSY;
+ }
+ if (!cxl_ops->link_ok(adapter, NULL))
+ return -EIO;
+ cpu_relax();
+ }
+ return 0;
+}
+
int cxl_invalidate_all_psl8(struct cxl *adapter)
{
unsigned long timeout = jiffies + (HZ * CXL_TIMEOUT);
@@ -553,10 +600,19 @@ static u64 calculate_sr(struct cxl_context *ctx)
sr |= (mfmsr() & MSR_SF) | CXL_PSL_SR_An_HV;
} else {
sr |= CXL_PSL_SR_An_PR | CXL_PSL_SR_An_R;
- sr &= ~(CXL_PSL_SR_An_HV);
+ if (radix_enabled())
+ sr |= CXL_PSL_SR_An_HV;
+ else
+ sr &= ~(CXL_PSL_SR_An_HV);
if (!test_tsk_thread_flag(current, TIF_32BIT))
sr |= CXL_PSL_SR_An_SF;
}
+ if (cxl_is_psl9(ctx->afu)) {
+ if (radix_enabled())
+ sr |= CXL_PSL_SR_An_XLAT_ror;
+ else
+ sr |= CXL_PSL_SR_An_XLAT_hpt;
+ }
return sr;
}
@@ -589,6 +645,70 @@ static void update_ivtes_directed(struct cxl_context *ctx)
WARN_ON(add_process_element(ctx));
}
+static int process_element_entry(struct cxl_context *ctx, u64 wed, u64 amr)
+{
+ u32 pid;
+
+ cxl_assign_psn_space(ctx);
+
+ ctx->elem->ctxtime = 0; /* disable */
+ ctx->elem->lpid = cpu_to_be32(mfspr(SPRN_LPID));
+ ctx->elem->haurp = 0; /* disable */
+
+ if (ctx->kernel)
+ pid = 0;
+ else {
+ if (ctx->mm == NULL) {
+ pr_devel("%s: unable to get mm for pe=%d pid=%i\n",
+ __func__, ctx->pe, pid_nr(ctx->pid));
+ return -EINVAL;
+ }
+ pid = ctx->mm->context.id;
+ }
+
+ ctx->elem->common.tid = 0;
+ ctx->elem->common.pid = cpu_to_be32(pid);
+
+ ctx->elem->sr = cpu_to_be64(calculate_sr(ctx));
+
+ ctx->elem->common.csrp = 0; /* disable */
+
+ cxl_prefault(ctx, wed);
+
+ /*
+ * Ensure we have the multiplexed PSL interrupt set up to take faults
+ * for kernel contexts that may not have allocated any AFU IRQs at all:
+ */
+ if (ctx->irqs.range[0] == 0) {
+ ctx->irqs.offset[0] = ctx->afu->native->psl_hwirq;
+ ctx->irqs.range[0] = 1;
+ }
+
+ ctx->elem->common.amr = cpu_to_be64(amr);
+ ctx->elem->common.wed = cpu_to_be64(wed);
+
+ return 0;
+}
+
+int cxl_attach_afu_directed_psl9(struct cxl_context *ctx, u64 wed, u64 amr)
+{
+ int result;
+
+ /* fill the process element entry */
+ result = process_element_entry(ctx, wed, amr);
+ if (result)
+ return result;
+
+ update_ivtes_directed(ctx);
+
+ /* first guy needs to enable */
+ result = cxl_ops->afu_check_and_enable(ctx->afu);
+ if (result)
+ return result;
+
+ return add_process_element(ctx);
+}
+
int cxl_attach_afu_directed_psl8(struct cxl_context *ctx, u64 wed, u64 amr)
{
u32 pid;
@@ -599,7 +719,7 @@ int cxl_attach_afu_directed_psl8(struct cxl_context *ctx, u64 wed, u64 amr)
ctx->elem->ctxtime = 0; /* disable */
ctx->elem->lpid = cpu_to_be32(mfspr(SPRN_LPID));
ctx->elem->haurp = 0; /* disable */
- ctx->elem->sdr = cpu_to_be64(mfspr(SPRN_SDR1));
+ ctx->elem->u.sdr = cpu_to_be64(mfspr(SPRN_SDR1));
pid = current->pid;
if (ctx->kernel)
@@ -610,13 +730,13 @@ int cxl_attach_afu_directed_psl8(struct cxl_context *ctx, u64 wed, u64 amr)
ctx->elem->sr = cpu_to_be64(calculate_sr(ctx));
ctx->elem->common.csrp = 0; /* disable */
- ctx->elem->common.aurp0 = 0; /* disable */
- ctx->elem->common.aurp1 = 0; /* disable */
+ ctx->elem->common.u.psl8.aurp0 = 0; /* disable */
+ ctx->elem->common.u.psl8.aurp1 = 0; /* disable */
cxl_prefault(ctx, wed);
- ctx->elem->common.sstp0 = cpu_to_be64(ctx->sstp0);
- ctx->elem->common.sstp1 = cpu_to_be64(ctx->sstp1);
+ ctx->elem->common.u.psl8.sstp0 = cpu_to_be64(ctx->sstp0);
+ ctx->elem->common.u.psl8.sstp1 = cpu_to_be64(ctx->sstp1);
/*
* Ensure we have the multiplexed PSL interrupt set up to take faults
@@ -682,6 +802,32 @@ static int deactivate_afu_directed(struct cxl_afu *afu)
return 0;
}
+int cxl_activate_dedicated_process_psl9(struct cxl_afu *afu)
+{
+ dev_info(&afu->dev, "Activating dedicated process mode\n");
+
+ /* If XSL is set to dedicated mode (Set in PSL_SCNTL reg), the
+ * XSL and AFU are programmed to work with a single context.
+ * The context information should be configured in the SPA area
+ * index 0 (so PSL_SPAP must be configured before enabling the
+ * AFU).
+ */
+ afu->num_procs = 1;
+ if (afu->native->spa == NULL) {
+ if (cxl_alloc_spa(afu))
+ return -ENOMEM;
+ }
+ attach_spa(afu);
+
+ cxl_p1n_write(afu, CXL_PSL_SCNTL_An, CXL_PSL_SCNTL_An_PM_Process);
+ cxl_p1n_write(afu, CXL_PSL_ID_An, CXL_PSL_ID_An_F | CXL_PSL_ID_An_L);
+
+ afu->current_mode = CXL_MODE_DEDICATED;
+ afu->num_procs = 1;
+
+ return cxl_chardev_d_afu_add(afu);
+}
+
int cxl_activate_dedicated_process_psl8(struct cxl_afu *afu)
{
dev_info(&afu->dev, "Activating dedicated process mode\n");
@@ -705,6 +851,16 @@ int cxl_activate_dedicated_process_psl8(struct cxl_afu *afu)
return cxl_chardev_d_afu_add(afu);
}
+void cxl_update_dedicated_ivtes_psl9(struct cxl_context *ctx)
+{
+ int r;
+
+ for (r = 0; r < CXL_IRQ_RANGES; r++) {
+ ctx->elem->ivte_offsets[r] = cpu_to_be16(ctx->irqs.offset[r]);
+ ctx->elem->ivte_ranges[r] = cpu_to_be16(ctx->irqs.range[r]);
+ }
+}
+
void cxl_update_dedicated_ivtes_psl8(struct cxl_context *ctx)
{
struct cxl_afu *afu = ctx->afu;
@@ -721,6 +877,26 @@ void cxl_update_dedicated_ivtes_psl8(struct cxl_context *ctx)
((u64)ctx->irqs.range[3] & 0xffff));
}
+int cxl_attach_dedicated_process_psl9(struct cxl_context *ctx, u64 wed, u64 amr)
+{
+ struct cxl_afu *afu = ctx->afu;
+ int result;
+
+ /* fill the process element entry */
+ result = process_element_entry(ctx, wed, amr);
+ if (result)
+ return result;
+
+ if (ctx->afu->adapter->native->sl_ops->update_dedicated_ivtes)
+ afu->adapter->native->sl_ops->update_dedicated_ivtes(ctx);
+
+ result = cxl_ops->afu_reset(afu);
+ if (result)
+ return result;
+
+ return afu_enable(afu);
+}
+
int cxl_attach_dedicated_process_psl8(struct cxl_context *ctx, u64 wed, u64 amr)
{
struct cxl_afu *afu = ctx->afu;
@@ -892,6 +1068,21 @@ static int native_get_irq_info(struct cxl_afu *afu, struct cxl_irq_info *info)
return 0;
}
+void cxl_native_irq_dump_regs_psl9(struct cxl_context *ctx)
+{
+ u64 fir1, fir2, serr;
+
+ fir1 = cxl_p1_read(ctx->afu->adapter, CXL_PSL9_FIR1);
+ fir2 = cxl_p1_read(ctx->afu->adapter, CXL_PSL9_FIR2);
+
+ dev_crit(&ctx->afu->dev, "PSL_FIR1: 0x%016llx\n", fir1);
+ dev_crit(&ctx->afu->dev, "PSL_FIR2: 0x%016llx\n", fir2);
+ if (ctx->afu->adapter->native->sl_ops->register_serr_irq) {
+ serr = cxl_p1n_read(ctx->afu, CXL_PSL_SERR_An);
+ cxl_afu_decode_psl_serr(ctx->afu, serr);
+ }
+}
+
void cxl_native_irq_dump_regs_psl8(struct cxl_context *ctx)
{
u64 fir1, fir2, fir_slice, serr, afu_debug;
@@ -928,6 +1119,16 @@ static irqreturn_t native_handle_psl_slice_error(struct cxl_context *ctx,
return cxl_ops->ack_irq(ctx, 0, errstat);
}
+irqreturn_t cxl_fail_irq_psl9(struct cxl_afu *afu, struct cxl_irq_info *irq_info)
+{
+ if (irq_info->dsisr & CXL_PSL9_DSISR_An_TF)
+ cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_AE);
+ else
+ cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_A);
+
+ return IRQ_HANDLED;
+}
+
irqreturn_t cxl_fail_irq_psl8(struct cxl_afu *afu, struct cxl_irq_info *irq_info)
{
if (irq_info->dsisr & CXL_PSL_DSISR_TRANS)
@@ -999,6 +1200,9 @@ static void native_irq_wait(struct cxl_context *ctx)
if (cxl_is_psl8(ctx->afu) &&
((dsisr & CXL_PSL_DSISR_PENDING) == 0))
return;
+ if (cxl_is_psl9(ctx->afu) &&
+ ((dsisr & CXL_PSL9_DSISR_PENDING) == 0))
+ return;
/*
* We are waiting for the workqueue to process our
* irq, so need to let that run here.
@@ -1127,6 +1331,8 @@ int cxl_native_register_serr_irq(struct cxl_afu *afu)
serr = cxl_p1n_read(afu, CXL_PSL_SERR_An);
if (cxl_is_power8())
serr = (serr & 0x00ffffffffff0000ULL) | (afu->serr_hwirq & 0xffff);
+ if (cxl_is_power9())
+ serr = (serr & ~0x0000000075010000ULL) | (afu->serr_hwirq & 0xffff);
cxl_p1n_write(afu, CXL_PSL_SERR_An, serr);
return 0;
diff --git a/drivers/misc/cxl/pci.c b/drivers/misc/cxl/pci.c
index 5880092..c9f6215 100644
--- a/drivers/misc/cxl/pci.c
+++ b/drivers/misc/cxl/pci.c
@@ -60,7 +60,7 @@
#define CXL_VSEC_PROTOCOL_MASK 0xe0
#define CXL_VSEC_PROTOCOL_1024TB 0x80
#define CXL_VSEC_PROTOCOL_512TB 0x40
-#define CXL_VSEC_PROTOCOL_256TB 0x20 /* Power 8 uses this */
+#define CXL_VSEC_PROTOCOL_256TB 0x20 /* Power 8/9 uses this */
#define CXL_VSEC_PROTOCOL_ENABLE 0x01
#define CXL_READ_VSEC_PSL_REVISION(dev, vsec, dest) \
@@ -326,14 +326,20 @@ static void dump_afu_descriptor(struct cxl_afu *afu)
#define P8_CAPP_UNIT0_ID 0xBA
#define P8_CAPP_UNIT1_ID 0XBE
+#define P9_CAPP_UNIT0_ID 0xC0
+#define P9_CAPP_UNIT1_ID 0xE0
-static u64 get_capp_unit_id(struct device_node *np)
+static u32 get_phb_index(struct device_node *np)
{
u32 phb_index;
if (of_property_read_u32(np, "ibm,phb-index", &phb_index))
- return 0;
+ return -ENODEV;
+ return phb_index;
+}
+static u64 get_capp_unit_id(struct device_node *np, u32 phb_index)
+{
/*
* POWER 8:
* - For chips other than POWER8NVL, we only have CAPP 0,
@@ -352,10 +358,25 @@ static u64 get_capp_unit_id(struct device_node *np)
return P8_CAPP_UNIT1_ID;
}
+ /*
+ * POWER 9:
+ * PEC0 (PHB0). Capp ID = CAPP0 (0b1100_0000)
+ * PEC1 (PHB1 - PHB2). No capi mode
+ * PEC2 (PHB3 - PHB4 - PHB5): Capi mode on PHB3 only. Capp ID = CAPP1 (0b1110_0000)
+ */
+ if (cxl_is_power9()) {
+ if (phb_index == 0)
+ return P9_CAPP_UNIT0_ID;
+
+ if (phb_index == 3)
+ return P9_CAPP_UNIT1_ID;
+ }
+
return 0;
}
-static int calc_capp_routing(struct pci_dev *dev, u64 *chipid, u64 *capp_unit_id)
+static int calc_capp_routing(struct pci_dev *dev, u64 *chipid,
+ u32 *phb_index, u64 *capp_unit_id)
{
struct device_node *np;
const __be32 *prop;
@@ -367,8 +388,16 @@ static int calc_capp_routing(struct pci_dev *dev, u64 *chipid, u64 *capp_unit_id
np = of_get_next_parent(np);
if (!np)
return -ENODEV;
+
*chipid = be32_to_cpup(prop);
- *capp_unit_id = get_capp_unit_id(np);
+
+ *phb_index = get_phb_index(np);
+ if (*phb_index == -ENODEV) {
+ pr_err("cxl: invalid phb index\n");
+ return -ENODEV;
+ }
+
+ *capp_unit_id = get_capp_unit_id(np, *phb_index);
of_node_put(np);
if (!*capp_unit_id) {
pr_err("cxl: invalid capp unit id\n");
@@ -378,14 +407,99 @@ static int calc_capp_routing(struct pci_dev *dev, u64 *chipid, u64 *capp_unit_id
return 0;
}
+static int init_implementation_adapter_regs_psl9(struct cxl *adapter, struct pci_dev *dev)
+{
+ u64 xsl_dsnctl, psl_fircntl;
+ u64 chipid;
+ u32 phb_index;
+ u64 capp_unit_id;
+ int rc;
+
+ rc = calc_capp_routing(dev, &chipid, &phb_index, &capp_unit_id);
+ if (rc)
+ return rc;
+
+ /* CAPI Identifier bits [0:7]
+ * bit 61:60 MSI bits --> 0
+ * bit 59 TVT selector --> 0
+ */
+ /* Tell XSL where to route data to.
+ * The field chipid should match the PHB CAPI_CMPM register
+ */
+ xsl_dsnctl = ((u64)0x2 << (63-7)); /* Bit 57 */
+ xsl_dsnctl |= (capp_unit_id << (63-15));
+
+ /* nMMU_ID Defaults to: b’000001001’*/
+ xsl_dsnctl |= ((u64)0x09 << (63-28));
+
+ /* Used to identify CAPI packets which should be sorted into
+ * the Non-Blocking queues by the PHB. This field should match
+ * the PHB PBL_NBW_CMPM register
+ * nbwind=0x03, bits [57:58], must include capi indicator
+ */
+ xsl_dsnctl |= ((u64)0x03 << (63-47));
+
+ /* Upper 16b address bits of ASB_Notify messages sent to the
+ * system. Need to match the PHB’s ASN Compare/Mask Register.
+ */
+ xsl_dsnctl |= ((u64)0x04 << (63-55));
+
+ cxl_p1_write(adapter, CXL_XSL9_DSNCTL, xsl_dsnctl);
+
+ /* Set fir_cntl to recommended value for production env */
+ psl_fircntl = (0x2ULL << (63-3)); /* ce_report */
+ psl_fircntl |= (0x1ULL << (63-6)); /* FIR_report */
+ psl_fircntl |= 0x1ULL; /* ce_thresh */
+ cxl_p1_write(adapter, CXL_PSL9_FIR_CNTL, psl_fircntl);
+
+ /* vccredits=0x1 pcklat=0x4 */
+ cxl_p1_write(adapter, CXL_PSL9_DSNDCTL, 0x0000000000001810ULL);
+
+ /* For debugging with trace arrays.
+ * Configure RX trace 0 segmented mode.
+ * Configure CT trace 0 segmented mode.
+ * Configure LA0 trace 0 segmented mode.
+ * Configure LA1 trace 0 segmented mode.
+ */
+ cxl_p1_write(adapter, CXL_PSL9_TRACECFG, 0x8040800080000000ULL);
+ cxl_p1_write(adapter, CXL_PSL9_TRACECFG, 0x8040800080000003ULL);
+ cxl_p1_write(adapter, CXL_PSL9_TRACECFG, 0x8040800080000005ULL);
+ cxl_p1_write(adapter, CXL_PSL9_TRACECFG, 0x8040800080000006ULL);
+
+ /* A response to an ASB_Notify request is returned by the
+ * system as an MMIO write to the address defined in
+ * the PSL_TNR_ADDR register
+ */
+ /* PSL_TNR_ADDR */
+
+ /* NORST */
+ cxl_p1_write(adapter, CXL_PSL9_DEBUG, 0x8000000000000000ULL);
+
+ /* allocate the apc machines. For PHB0, let us keep the APC
+ * allocation setup as it is. For PHB3, we need to disable Rd
+ * machines
+ */
+ if (phb_index == 3) {
+ cxl_p1_write(adapter, CXL_PSL9_APCDEDALLOC, 0x8000808200000000ULL);
+ cxl_p1_write(adapter, CXL_PSL9_APCDEDTYPE, 0x40000003FFFF0000ULL);
+ }
+
+ /* Disable vc dd1 fix */
+ if ((cxl_is_power9() && cpu_has_feature(CPU_FTR_POWER9_DD1)))
+ cxl_p1_write(adapter, CXL_PSL9_GP_CT, 0x0400000000000001ULL);
+
+ return 0;
+}
+
static int init_implementation_adapter_regs_psl8(struct cxl *adapter, struct pci_dev *dev)
{
u64 psl_dsnctl, psl_fircntl;
u64 chipid;
+ u32 phb_index;
u64 capp_unit_id;
int rc;
- rc = calc_capp_routing(dev, &chipid, &capp_unit_id);
+ rc = calc_capp_routing(dev, &chipid, &phb_index, &capp_unit_id);
if (rc)
return rc;
@@ -414,10 +528,11 @@ static int init_implementation_adapter_regs_xsl(struct cxl *adapter, struct pci_
{
u64 xsl_dsnctl;
u64 chipid;
+ u32 phb_index;
u64 capp_unit_id;
int rc;
- rc = calc_capp_routing(dev, &chipid, &capp_unit_id);
+ rc = calc_capp_routing(dev, &chipid, &phb_index, &capp_unit_id);
if (rc)
return rc;
@@ -435,6 +550,12 @@ static int init_implementation_adapter_regs_xsl(struct cxl *adapter, struct pci_
/* For the PSL this is a multiple for 0 < n <= 7: */
#define PSL_2048_250MHZ_CYCLES 1
+static void write_timebase_ctrl_psl9(struct cxl *adapter)
+{
+ cxl_p1_write(adapter, CXL_PSL9_TB_CTLSTAT,
+ TBSYNC_CNT(2 * PSL_2048_250MHZ_CYCLES));
+}
+
static void write_timebase_ctrl_psl8(struct cxl *adapter)
{
cxl_p1_write(adapter, CXL_PSL_TB_CTLSTAT,
@@ -456,6 +577,11 @@ static void write_timebase_ctrl_xsl(struct cxl *adapter)
TBSYNC_CNT(XSL_4000_CLOCKS));
}
+static u64 timebase_read_psl9(struct cxl *adapter)
+{
+ return cxl_p1_read(adapter, CXL_PSL9_Timebase);
+}
+
static u64 timebase_read_psl8(struct cxl *adapter)
{
return cxl_p1_read(adapter, CXL_PSL_Timebase);
@@ -514,6 +640,11 @@ static void cxl_setup_psl_timebase(struct cxl *adapter, struct pci_dev *dev)
return;
}
+static int init_implementation_afu_regs_psl9(struct cxl_afu *afu)
+{
+ return 0;
+}
+
static int init_implementation_afu_regs_psl8(struct cxl_afu *afu)
{
/* read/write masks for this slice */
@@ -612,7 +743,7 @@ static int setup_cxl_bars(struct pci_dev *dev)
/*
* BAR 4/5 has a special meaning for CXL and must be programmed with a
* special value corresponding to the CXL protocol address range.
- * For POWER 8 that means bits 48:49 must be set to 10
+ * For POWER 8/9 that means bits 48:49 must be set to 10
*/
pci_write_config_dword(dev, PCI_BASE_ADDRESS_4, 0x00000000);
pci_write_config_dword(dev, PCI_BASE_ADDRESS_5, 0x00020000);
@@ -997,6 +1128,52 @@ static int cxl_afu_descriptor_looks_ok(struct cxl_afu *afu)
return 0;
}
+static int sanitise_afu_regs_psl9(struct cxl_afu *afu)
+{
+ u64 reg;
+
+ /*
+ * Clear out any regs that contain either an IVTE or address or may be
+ * waiting on an acknowledgment to try to be a bit safer as we bring
+ * it online
+ */
+ reg = cxl_p2n_read(afu, CXL_AFU_Cntl_An);
+ if ((reg & CXL_AFU_Cntl_An_ES_MASK) != CXL_AFU_Cntl_An_ES_Disabled) {
+ dev_warn(&afu->dev, "WARNING: AFU was not disabled: %#016llx\n", reg);
+ if (cxl_ops->afu_reset(afu))
+ return -EIO;
+ if (cxl_afu_disable(afu))
+ return -EIO;
+ if (cxl_psl_purge(afu))
+ return -EIO;
+ }
+ cxl_p1n_write(afu, CXL_PSL_SPAP_An, 0x0000000000000000);
+ cxl_p1n_write(afu, CXL_PSL_AMBAR_An, 0x0000000000000000);
+ reg = cxl_p2n_read(afu, CXL_PSL_DSISR_An);
+ if (reg) {
+ dev_warn(&afu->dev, "AFU had pending DSISR: %#016llx\n", reg);
+ if (reg & CXL_PSL9_DSISR_An_TF)
+ cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_AE);
+ else
+ cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_A);
+ }
+ if (afu->adapter->native->sl_ops->register_serr_irq) {
+ reg = cxl_p1n_read(afu, CXL_PSL_SERR_An);
+ if (reg) {
+ if (reg & ~0x000000007501ffff)
+ dev_warn(&afu->dev, "AFU had pending SERR: %#016llx\n", reg);
+ cxl_p1n_write(afu, CXL_PSL_SERR_An, reg & ~0xffff);
+ }
+ }
+ reg = cxl_p2n_read(afu, CXL_PSL_ErrStat_An);
+ if (reg) {
+ dev_warn(&afu->dev, "AFU had pending error status: %#016llx\n", reg);
+ cxl_p2n_write(afu, CXL_PSL_ErrStat_An, reg);
+ }
+
+ return 0;
+}
+
static int sanitise_afu_regs_psl8(struct cxl_afu *afu)
{
u64 reg;
@@ -1393,6 +1570,9 @@ static bool cxl_compatible_caia_version(struct cxl *adapter)
if (cxl_is_power8() && (adapter->caia_major == 1))
return true;
+ if (cxl_is_power9() && (adapter->caia_major == 2))
+ return true;
+
return false;
}
@@ -1546,6 +1726,30 @@ static void cxl_deconfigure_adapter(struct cxl *adapter)
pci_disable_device(pdev);
}
+static const struct cxl_service_layer_ops psl9_ops = {
+ .adapter_regs_init = init_implementation_adapter_regs_psl9,
+ .invalidate_all = cxl_invalidate_all_psl9,
+ .afu_regs_init = init_implementation_afu_regs_psl9,
+ .sanitise_afu_regs = sanitise_afu_regs_psl9,
+ .register_serr_irq = cxl_native_register_serr_irq,
+ .release_serr_irq = cxl_native_release_serr_irq,
+ .handle_interrupt = cxl_irq_psl9,
+ .fail_irq = cxl_fail_irq_psl9,
+ .activate_dedicated_process = cxl_activate_dedicated_process_psl9,
+ .attach_afu_directed = cxl_attach_afu_directed_psl9,
+ .attach_dedicated_process = cxl_attach_dedicated_process_psl9,
+ .update_dedicated_ivtes = cxl_update_dedicated_ivtes_psl9,
+ .debugfs_add_adapter_regs = cxl_debugfs_add_adapter_regs_psl9,
+ .debugfs_add_afu_regs = cxl_debugfs_add_afu_regs_psl9,
+ .psl_irq_dump_registers = cxl_native_irq_dump_regs_psl9,
+ .err_irq_dump_registers = cxl_native_err_irq_dump_regs,
+ .debugfs_stop_trace = cxl_stop_trace_psl9,
+ .write_timebase_ctrl = write_timebase_ctrl_psl9,
+ .timebase_read = timebase_read_psl9,
+ .capi_mode = OPAL_PHB_CAPI_MODE_CAPI,
+ .needs_reset_before_disable = true,
+};
+
static const struct cxl_service_layer_ops psl8_ops = {
.adapter_regs_init = init_implementation_adapter_regs_psl8,
.invalidate_all = cxl_invalidate_all_psl8,
@@ -1589,6 +1793,9 @@ static void set_sl_ops(struct cxl *adapter, struct pci_dev *dev)
if (cxl_is_power8()) {
dev_info(&dev->dev, "Device uses a PSL8\n");
adapter->native->sl_ops = &psl8_ops;
+ } else {
+ dev_info(&dev->dev, "Device uses a PSL9\n");
+ adapter->native->sl_ops = &psl9_ops;
}
}
}
@@ -1744,6 +1951,11 @@ static int cxl_probe(struct pci_dev *dev, const struct pci_device_id *id)
return -ENODEV;
}
+ if (cxl_is_power9() && !radix_enabled()) {
+ dev_info(&dev->dev, "Only Radix mode supported\n");
+ return -ENODEV;
+ }
+
if (cxl_verbose)
dump_cxl_config_space(dev);
diff --git a/drivers/misc/cxl/trace.h b/drivers/misc/cxl/trace.h
index 751d611..b8e300a 100644
--- a/drivers/misc/cxl/trace.h
+++ b/drivers/misc/cxl/trace.h
@@ -17,6 +17,15 @@
#include "cxl.h"
+#define dsisr_psl9_flags(flags) \
+ __print_flags(flags, "|", \
+ { CXL_PSL9_DSISR_An_CO_MASK, "FR" }, \
+ { CXL_PSL9_DSISR_An_TF, "TF" }, \
+ { CXL_PSL9_DSISR_An_PE, "PE" }, \
+ { CXL_PSL9_DSISR_An_AE, "AE" }, \
+ { CXL_PSL9_DSISR_An_OC, "OC" }, \
+ { CXL_PSL9_DSISR_An_S, "S" })
+
#define DSISR_FLAGS \
{ CXL_PSL_DSISR_An_DS, "DS" }, \
{ CXL_PSL_DSISR_An_DM, "DM" }, \
@@ -154,6 +163,40 @@ TRACE_EVENT(cxl_afu_irq,
)
);
+TRACE_EVENT(cxl_psl9_irq,
+ TP_PROTO(struct cxl_context *ctx, int irq, u64 dsisr, u64 dar),
+
+ TP_ARGS(ctx, irq, dsisr, dar),
+
+ TP_STRUCT__entry(
+ __field(u8, card)
+ __field(u8, afu)
+ __field(u16, pe)
+ __field(int, irq)
+ __field(u64, dsisr)
+ __field(u64, dar)
+ ),
+
+ TP_fast_assign(
+ __entry->card = ctx->afu->adapter->adapter_num;
+ __entry->afu = ctx->afu->slice;
+ __entry->pe = ctx->pe;
+ __entry->irq = irq;
+ __entry->dsisr = dsisr;
+ __entry->dar = dar;
+ ),
+
+ TP_printk("afu%i.%i pe=%i irq=%i dsisr=0x%016llx dsisr=%s dar=0x%016llx",
+ __entry->card,
+ __entry->afu,
+ __entry->pe,
+ __entry->irq,
+ __entry->dsisr,
+ dsisr_psl9_flags(__entry->dsisr),
+ __entry->dar
+ )
+);
+
TRACE_EVENT(cxl_psl_irq,
TP_PROTO(struct cxl_context *ctx, int irq, u64 dsisr, u64 dar),
--
2.7.4
^ permalink raw reply related [flat|nested] 20+ messages in thread* Re: [PATCH V2 6/6] cxl: Add psl9 specific code
2017-03-14 11:08 ` [PATCH V2 6/6] cxl: Add psl9 specific code Christophe Lombard
@ 2017-03-21 11:21 ` Frederic Barrat
0 siblings, 0 replies; 20+ messages in thread
From: Frederic Barrat @ 2017-03-21 11:21 UTC (permalink / raw)
To: Christophe Lombard, linuxppc-dev, imunsie, andrew.donnellan
Le 14/03/2017 à 12:08, Christophe Lombard a écrit :
> The new Coherent Accelerator Interface Architecture, level 2, for the
> IBM POWER9 brings new content and features:
> - POWER9 Service Layer
> - Registers
> - Radix mode
> - Process element entry
> - Dedicated-Shared Process Programming Model
> - Translation Fault Handling
> - CAPP
> - Memory Context ID
> If a valid mm_struct is found the memory context id is used for each
> transaction associated with the process handle. The PSL uses the
> context ID to find the corresponding process element.
>
> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com>
> ---
> drivers/misc/cxl/context.c | 13 +++
> drivers/misc/cxl/cxl.h | 124 ++++++++++++++++++++----
> drivers/misc/cxl/debugfs.c | 19 ++++
> drivers/misc/cxl/fault.c | 48 ++++++----
> drivers/misc/cxl/guest.c | 8 +-
> drivers/misc/cxl/irq.c | 53 +++++++++++
> drivers/misc/cxl/native.c | 218 +++++++++++++++++++++++++++++++++++++++++--
> drivers/misc/cxl/pci.c | 228 +++++++++++++++++++++++++++++++++++++++++++--
> drivers/misc/cxl/trace.h | 43 +++++++++
> 9 files changed, 700 insertions(+), 54 deletions(-)
>
> diff --git a/drivers/misc/cxl/context.c b/drivers/misc/cxl/context.c
> index 0fd3cf8..8d41fc5 100644
> --- a/drivers/misc/cxl/context.c
> +++ b/drivers/misc/cxl/context.c
> @@ -205,6 +205,19 @@ int cxl_context_iomap(struct cxl_context *ctx, struct vm_area_struct *vma)
> return -EBUSY;
> }
>
> + if ((ctx->afu->current_mode == CXL_MODE_DEDICATED) &&
> + (cxl_is_psl9(ctx->afu))) {
> + /* make sure there is a valid problem state area space for this AFU */
> + if (ctx->master && !ctx->afu->psa) {
> + pr_devel("AFU doesn't support mmio space\n");
> + return -EINVAL;
> + }
> +
> + /* Can't mmap until the AFU is enabled */
> + if (!ctx->afu->enabled)
> + return -EBUSY;
> + }
> +
It looks we could refactor the if clause with the above...
> pr_devel("%s: mmio physical: %llx pe: %i master:%i\n", __func__,
> ctx->psn_phys, ctx->pe , ctx->master);
>
> diff --git a/drivers/misc/cxl/cxl.h b/drivers/misc/cxl/cxl.h
> index f6a3a34..fbdc511 100644
> --- a/drivers/misc/cxl/cxl.h
> +++ b/drivers/misc/cxl/cxl.h
> @@ -63,7 +63,7 @@ typedef struct {
> /* Memory maps. Ref CXL Appendix A */
>
> /* PSL Privilege 1 Memory Map */
> -/* Configuration and Control area */
> +/* Configuration and Control area - CAIA 1&2 */
> static const cxl_p1_reg_t CXL_PSL_CtxTime = {0x0000};
> static const cxl_p1_reg_t CXL_PSL_ErrIVTE = {0x0008};
> static const cxl_p1_reg_t CXL_PSL_KEY1 = {0x0010};
> @@ -98,11 +98,29 @@ static const cxl_p1_reg_t CXL_XSL_Timebase = {0x0100};
> static const cxl_p1_reg_t CXL_XSL_TB_CTLSTAT = {0x0108};
> static const cxl_p1_reg_t CXL_XSL_FEC = {0x0158};
> static const cxl_p1_reg_t CXL_XSL_DSNCTL = {0x0168};
> +/* PSL registers - CAIA 2 */
> +static const cxl_p1_reg_t CXL_PSL9_CONTROL = {0x0020};
> +static const cxl_p1_reg_t CXL_XSL9_DSNCTL = {0x0168};
> +static const cxl_p1_reg_t CXL_PSL9_FIR1 = {0x0300};
> +static const cxl_p1_reg_t CXL_PSL9_FIR2 = {0x0308}; /* TBD NML CAIA 2 */
TBD NML???
> +static const cxl_p1_reg_t CXL_PSL9_Timebase = {0x0310};
> +static const cxl_p1_reg_t CXL_PSL9_DEBUG = {0x0320};
> +static const cxl_p1_reg_t CXL_PSL9_FIR_CNTL = {0x0348};
> +static const cxl_p1_reg_t CXL_PSL9_DSNDCTL = {0x0350};
> +static const cxl_p1_reg_t CXL_PSL9_TB_CTLSTAT = {0x0340};
> +static const cxl_p1_reg_t CXL_PSL9_TRACECFG = {0x0368};
> +static const cxl_p1_reg_t CXL_PSL9_APCDEDALLOC = {0x0378};
> +static const cxl_p1_reg_t CXL_PSL9_APCDEDTYPE = {0x0380};
> +static const cxl_p1_reg_t CXL_PSL9_TNR_ADDR = {0x0388};
> +static const cxl_p1_reg_t CXL_PSL9_GP_CT = {0x0398};
> +static const cxl_p1_reg_t CXL_XSL9_IERAT = {0x0588};
> +static const cxl_p1_reg_t CXL_XSL9_ILPP = {0x0590};
> +
> /* 0x7F00:7FFF Reserved PCIe MSI-X Pending Bit Array area */
> /* 0x8000:FFFF Reserved PCIe MSI-X Table Area */
>
> /* PSL Slice Privilege 1 Memory Map */
> -/* Configuration Area */
> +/* Configuration Area - CAIA 1&2 */
> static const cxl_p1n_reg_t CXL_PSL_SR_An = {0x00};
> static const cxl_p1n_reg_t CXL_PSL_LPID_An = {0x08};
> static const cxl_p1n_reg_t CXL_PSL_AMBAR_An = {0x10};
> @@ -111,17 +129,18 @@ static const cxl_p1n_reg_t CXL_PSL_ID_An = {0x20};
> static const cxl_p1n_reg_t CXL_PSL_SERR_An = {0x28};
> /* Memory Management and Lookaside Buffer Management - CAIA 1*/
> static const cxl_p1n_reg_t CXL_PSL_SDR_An = {0x30};
> +/* Memory Management and Lookaside Buffer Management - CAIA 1&2 */
> static const cxl_p1n_reg_t CXL_PSL_AMOR_An = {0x38};
> -/* Pointer Area */
> +/* Pointer Area - CAIA 1&2 */
> static const cxl_p1n_reg_t CXL_HAURP_An = {0x80};
> static const cxl_p1n_reg_t CXL_PSL_SPAP_An = {0x88};
> static const cxl_p1n_reg_t CXL_PSL_LLCMD_An = {0x90};
> -/* Control Area */
> +/* Control Area - CAIA 1&2 */
> static const cxl_p1n_reg_t CXL_PSL_SCNTL_An = {0xA0};
> static const cxl_p1n_reg_t CXL_PSL_CtxTime_An = {0xA8};
> static const cxl_p1n_reg_t CXL_PSL_IVTE_Offset_An = {0xB0};
> static const cxl_p1n_reg_t CXL_PSL_IVTE_Limit_An = {0xB8};
> -/* 0xC0:FF Implementation Dependent Area */
> +/* 0xC0:FF Implementation Dependent Area - CAIA 1&2 */
> static const cxl_p1n_reg_t CXL_PSL_FIR_SLICE_An = {0xC0};
> static const cxl_p1n_reg_t CXL_AFU_DEBUG_An = {0xC8};
> /* 0xC0:FF Implementation Dependent Area - CAIA 1 */
> @@ -131,7 +150,7 @@ static const cxl_p1n_reg_t CXL_PSL_RXCTL_A = {0xE0};
> static const cxl_p1n_reg_t CXL_PSL_SLICE_TRACE = {0xE8};
>
> /* PSL Slice Privilege 2 Memory Map */
> -/* Configuration and Control Area */
> +/* Configuration and Control Area - CAIA 1&2 */
> static const cxl_p2n_reg_t CXL_PSL_PID_TID_An = {0x000};
> static const cxl_p2n_reg_t CXL_CSRP_An = {0x008};
> /* Configuration and Control Area - CAIA 1 */
> @@ -145,17 +164,17 @@ static const cxl_p2n_reg_t CXL_PSL_AMR_An = {0x030};
> static const cxl_p2n_reg_t CXL_SLBIE_An = {0x040};
> static const cxl_p2n_reg_t CXL_SLBIA_An = {0x048};
> static const cxl_p2n_reg_t CXL_SLBI_Select_An = {0x050};
> -/* Interrupt Registers */
> +/* Interrupt Registers - CAIA 1&2 */
> static const cxl_p2n_reg_t CXL_PSL_DSISR_An = {0x060};
> static const cxl_p2n_reg_t CXL_PSL_DAR_An = {0x068};
> static const cxl_p2n_reg_t CXL_PSL_DSR_An = {0x070};
> static const cxl_p2n_reg_t CXL_PSL_TFC_An = {0x078};
> static const cxl_p2n_reg_t CXL_PSL_PEHandle_An = {0x080};
> static const cxl_p2n_reg_t CXL_PSL_ErrStat_An = {0x088};
> -/* AFU Registers */
> +/* AFU Registers - CAIA 1&2 */
> static const cxl_p2n_reg_t CXL_AFU_Cntl_An = {0x090};
> static const cxl_p2n_reg_t CXL_AFU_ERR_An = {0x098};
> -/* Work Element Descriptor */
> +/* Work Element Descriptor - CAIA 1&2 */
> static const cxl_p2n_reg_t CXL_PSL_WED_An = {0x0A0};
> /* 0x0C0:FFF Implementation Dependent Area */
>
> @@ -182,6 +201,10 @@ static const cxl_p2n_reg_t CXL_PSL_WED_An = {0x0A0};
> #define CXL_PSL_SR_An_SF MSR_SF /* 64bit */
> #define CXL_PSL_SR_An_TA (1ull << (63-1)) /* Tags active, GA1: 0 */
> #define CXL_PSL_SR_An_HV MSR_HV /* Hypervisor, GA1: 0 */
> +#define CXL_PSL_SR_An_XLAT_hpt (0ull << (63-6))/* Hashed page table (HPT) mode */
> +#define CXL_PSL_SR_An_XLAT_roh (2ull << (63-6))/* Radix on HPT mode */
> +#define CXL_PSL_SR_An_XLAT_ror (3ull << (63-6))/* Radix on Radix mode */
> +#define CXL_PSL_SR_An_BOT (1ull << (63-10)) /* Use the in-memory segment table */
> #define CXL_PSL_SR_An_PR MSR_PR /* Problem state, GA1: 1 */
> #define CXL_PSL_SR_An_ISL (1ull << (63-53)) /* Ignore Segment Large Page */
> #define CXL_PSL_SR_An_TC (1ull << (63-54)) /* Page Table secondary hash */
> @@ -298,12 +321,38 @@ static const cxl_p2n_reg_t CXL_PSL_WED_An = {0x0A0};
> #define CXL_PSL_DSISR_An_S DSISR_ISSTORE /* Access was afu_wr or afu_zero */
> #define CXL_PSL_DSISR_An_K DSISR_KEYFAULT /* Access not permitted by virtual page class key protection */
>
> +/****** CXL_PSL_DSISR_An - CAIA 2 ****************************************************/
> +#define CXL_PSL9_DSISR_An_TF (1ull << (63-3)) /* Translation fault */
> +#define CXL_PSL9_DSISR_An_PE (1ull << (63-4)) /* PSL Error (implementation specific) */
> +#define CXL_PSL9_DSISR_An_AE (1ull << (63-5)) /* AFU Error */
> +#define CXL_PSL9_DSISR_An_OC (1ull << (63-6)) /* OS Context Warning */
> +#define CXL_PSL9_DSISR_An_S (1ull << (63-38)) /* TF for a write operation */
> +#define CXL_PSL9_DSISR_PENDING (CXL_PSL9_DSISR_An_TF | CXL_PSL9_DSISR_An_PE | CXL_PSL9_DSISR_An_AE | CXL_PSL9_DSISR_An_OC)
> +/* NOTE: Bits 56:63 (Checkout Response Status) are valid when DSISR_An[TF] = 1
> + * Status (0:7) Encoding
> + */
> +#define CXL_PSL9_DSISR_An_CO_MASK 0x00000000000000ffULL
> +#define CXL_PSL9_DSISR_An_SF 0x0000000000000080ULL /* Segment Fault 0b10000000 */
> +#define CXL_PSL9_DSISR_An_PF_SLR 0x0000000000000088ULL /* PTE not found (Single Level Radix) 0b10001000 */
> +#define CXL_PSL9_DSISR_An_PF_RGC 0x000000000000008CULL /* PTE not found (Radix Guest (child)) 0b10001100 */
> +#define CXL_PSL9_DSISR_An_PF_RGP 0x0000000000000090ULL /* PTE not found (Radix Guest (parent)) 0b10010000 */
> +#define CXL_PSL9_DSISR_An_PF_HRH 0x0000000000000094ULL /* PTE not found (HPT/Radix Host) 0b10010100 */
> +#define CXL_PSL9_DSISR_An_PF_STEG 0x000000000000009CULL /* PTE not found (STEG VA) 0b10011100 */
> +
> /****** CXL_PSL_TFC_An ******************************************************/
> #define CXL_PSL_TFC_An_A (1ull << (63-28)) /* Acknowledge non-translation fault */
> #define CXL_PSL_TFC_An_C (1ull << (63-29)) /* Continue (abort transaction) */
> #define CXL_PSL_TFC_An_AE (1ull << (63-30)) /* Restart PSL with address error */
> #define CXL_PSL_TFC_An_R (1ull << (63-31)) /* Restart PSL transaction */
>
> +/****** CXL_XSL9_IERAT_ERAT - CAIA 2 **********************************/
> +#define CXL_XSL9_IERAT_MLPID (1ull << (63-0)) /* Match LPID */
> +#define CXL_XSL9_IERAT_MPID (1ull << (63-1)) /* Match PID */
> +#define CXL_XSL9_IERAT_PRS (1ull << (63-4)) /* PRS bit for Radix invalidations */
> +#define CXL_XSL9_IERAT_INVR (1ull << (63-3)) /* Invalidate Radix */
> +#define CXL_XSL9_IERAT_IALL (1ull << (63-8)) /* Invalidate All */
> +#define CXL_XSL9_IERAT_IINPROG (1ull << (63-63)) /* Invalidate in progress */
> +
> /* cxl_process_element->software_status */
> #define CXL_PE_SOFTWARE_STATE_V (1ul << (31 - 0)) /* Valid */
> #define CXL_PE_SOFTWARE_STATE_C (1ul << (31 - 29)) /* Complete */
> @@ -654,25 +703,38 @@ int cxl_pci_reset(struct cxl *adapter);
> void cxl_pci_release_afu(struct device *dev);
> ssize_t cxl_pci_read_adapter_vpd(struct cxl *adapter, void *buf, size_t len);
>
> -/* common == phyp + powernv */
> +/* common == phyp + powernv - CAIA 1&2 */
> struct cxl_process_element_common {
> __be32 tid;
> __be32 pid;
> __be64 csrp;
> - __be64 aurp0;
> - __be64 aurp1;
> - __be64 sstp0;
> - __be64 sstp1;
> + union {
> + struct {
> + __be64 aurp0;
> + __be64 aurp1;
> + __be64 sstp0;
> + __be64 sstp1;
> + } psl8; /* CAIA 1 */
> + struct {
> + u8 reserved2[8];
> + u8 reserved3[8];
> + u8 reserved4[8];
> + u8 reserved5[8];
> + } psl9; /* CAIA 2 */
> + } u;
> __be64 amr;
> - u8 reserved3[4];
> + u8 reserved6[4];
> __be64 wed;
> } __packed;
>
> -/* just powernv */
> +/* just powernv - CAIA 1&2 */
> struct cxl_process_element {
> __be64 sr;
> __be64 SPOffset;
> - __be64 sdr;
> + union {
> + __be64 sdr; /* CAIA 1 */
> + u8 reserved1[8]; /* CAIA 2 */
> + } u;
> __be64 haurp;
> __be32 ctxtime;
> __be16 ivte_offsets[4];
> @@ -761,6 +823,16 @@ static inline bool cxl_is_power8(void)
> return false;
> }
>
> +static inline bool cxl_is_power9(void)
> +{
> + /* intermediate solution */
> + if (!cxl_is_power8() &&
> + (cpu_has_feature(CPU_FTRS_POWER9) ||
> + cpu_has_feature(CPU_FTR_POWER9_DD1)))
> + return true;
> + return false;
> +}
> +
It looks like we can define power9 the same way it was done for power8
now, the missing definitions have been added.
> static inline bool cxl_is_psl8(struct cxl_afu *afu)
> {
> if (afu->adapter->caia_major == 1)
> @@ -768,6 +840,13 @@ static inline bool cxl_is_psl8(struct cxl_afu *afu)
> return false;
> }
>
> +static inline bool cxl_is_psl9(struct cxl_afu *afu)
> +{
> + if (afu->adapter->caia_major == 2)
> + return true;
> + return false;
> +}
> +
> ssize_t cxl_pci_afu_read_err_buffer(struct cxl_afu *afu, char *buf,
> loff_t off, size_t count);
>
> @@ -834,9 +913,13 @@ void afu_irq_name_free(struct cxl_context *ctx);
>
> #ifdef CONFIG_DEBUG_FS
>
> +int cxl_attach_afu_directed_psl9(struct cxl_context *ctx, u64 wed, u64 amr);
> int cxl_attach_afu_directed_psl8(struct cxl_context *ctx, u64 wed, u64 amr);
> +int cxl_activate_dedicated_process_psl9(struct cxl_afu *afu);
> int cxl_activate_dedicated_process_psl8(struct cxl_afu *afu);
> +int cxl_attach_dedicated_process_psl9(struct cxl_context *ctx, u64 wed, u64 amr);
> int cxl_attach_dedicated_process_psl8(struct cxl_context *ctx, u64 wed, u64 amr);
> +void cxl_update_dedicated_ivtes_psl9(struct cxl_context *ctx);
> void cxl_update_dedicated_ivtes_psl8(struct cxl_context *ctx);
>
> int cxl_debugfs_init(void);
> @@ -845,9 +928,12 @@ int cxl_debugfs_adapter_add(struct cxl *adapter);
> void cxl_debugfs_adapter_remove(struct cxl *adapter);
> int cxl_debugfs_afu_add(struct cxl_afu *afu);
> void cxl_debugfs_afu_remove(struct cxl_afu *afu);
> +void cxl_stop_trace_psl9(struct cxl *cxl);
> void cxl_stop_trace_psl8(struct cxl *cxl);
> +void cxl_debugfs_add_adapter_regs_psl9(struct cxl *adapter, struct dentry *dir);
> void cxl_debugfs_add_adapter_regs_psl8(struct cxl *adapter, struct dentry *dir);
> void cxl_debugfs_add_adapter_regs_xsl(struct cxl *adapter, struct dentry *dir);
> +void cxl_debugfs_add_afu_regs_psl9(struct cxl_afu *afu, struct dentry *dir);
> void cxl_debugfs_add_afu_regs_psl8(struct cxl_afu *afu, struct dentry *dir);
>
> #else /* CONFIG_DEBUG_FS */
> @@ -950,8 +1036,11 @@ struct cxl_irq_info {
> };
>
> void cxl_assign_psn_space(struct cxl_context *ctx);
> +int cxl_invalidate_all_psl9(struct cxl *adapter);
> int cxl_invalidate_all_psl8(struct cxl *adapter);
> +irqreturn_t cxl_irq_psl9(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info);
> irqreturn_t cxl_irq_psl8(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info);
> +irqreturn_t cxl_fail_irq_psl9(struct cxl_afu *afu, struct cxl_irq_info *irq_info);
> irqreturn_t cxl_fail_irq_psl8(struct cxl_afu *afu, struct cxl_irq_info *irq_info);
> int cxl_register_one_irq(struct cxl *adapter, irq_handler_t handler,
> void *cookie, irq_hw_number_t *dest_hwirq,
> @@ -963,6 +1052,7 @@ int cxl_data_cache_flush(struct cxl *adapter);
> int cxl_afu_disable(struct cxl_afu *afu);
> int cxl_psl_purge(struct cxl_afu *afu);
>
> +void cxl_native_irq_dump_regs_psl9(struct cxl_context *ctx);
> void cxl_native_irq_dump_regs_psl8(struct cxl_context *ctx);
> void cxl_native_err_irq_dump_regs(struct cxl *adapter);
> int cxl_pci_vphb_add(struct cxl_afu *afu);
> diff --git a/drivers/misc/cxl/debugfs.c b/drivers/misc/cxl/debugfs.c
> index 43a1a27..eae9d74 100644
> --- a/drivers/misc/cxl/debugfs.c
> +++ b/drivers/misc/cxl/debugfs.c
> @@ -15,6 +15,12 @@
>
> static struct dentry *cxl_debugfs;
>
> +void cxl_stop_trace_psl9(struct cxl *adapter)
> +{
> + /* Stop the trace */
> + cxl_p1_write(adapter, CXL_PSL9_TRACECFG, 0x4480000000000000ULL);
> +}
> +
> void cxl_stop_trace_psl8(struct cxl *adapter)
> {
> int slice;
> @@ -53,6 +59,14 @@ static struct dentry *debugfs_create_io_x64(const char *name, umode_t mode,
> (void __force *)value, &fops_io_x64);
> }
>
> +void cxl_debugfs_add_adapter_regs_psl9(struct cxl *adapter, struct dentry *dir)
> +{
> + debugfs_create_io_x64("fir1", S_IRUSR, dir, _cxl_p1_addr(adapter, CXL_PSL9_FIR1));
> + debugfs_create_io_x64("fir2", S_IRUSR, dir, _cxl_p1_addr(adapter, CXL_PSL9_FIR2));
> + debugfs_create_io_x64("fir_cntl", S_IRUSR, dir, _cxl_p1_addr(adapter, CXL_PSL9_FIR_CNTL));
> + debugfs_create_io_x64("trace", S_IRUSR | S_IWUSR, dir, _cxl_p1_addr(adapter, CXL_PSL9_TRACECFG));
> +}
> +
> void cxl_debugfs_add_adapter_regs_psl8(struct cxl *adapter, struct dentry *dir)
> {
> debugfs_create_io_x64("fir1", S_IRUSR, dir, _cxl_p1_addr(adapter, CXL_PSL_FIR1));
> @@ -92,6 +106,11 @@ void cxl_debugfs_adapter_remove(struct cxl *adapter)
> debugfs_remove_recursive(adapter->debugfs);
> }
>
> +void cxl_debugfs_add_afu_regs_psl9(struct cxl_afu *afu, struct dentry *dir)
> +{
> + debugfs_create_io_x64("serr", S_IRUSR, dir, _cxl_p1n_addr(afu, CXL_PSL_SERR_An));
> +}
> +
> void cxl_debugfs_add_afu_regs_psl8(struct cxl_afu *afu, struct dentry *dir)
> {
> debugfs_create_io_x64("sstp0", S_IRUSR, dir, _cxl_p2n_addr(afu, CXL_SSTP0_An));
> diff --git a/drivers/misc/cxl/fault.c b/drivers/misc/cxl/fault.c
> index 9b72b90..3f84960 100644
> --- a/drivers/misc/cxl/fault.c
> +++ b/drivers/misc/cxl/fault.c
> @@ -146,25 +146,26 @@ static void cxl_handle_page_fault(struct cxl_context *ctx,
> return cxl_ack_ae(ctx);
> }
>
> - /*
> - * update_mmu_cache() will not have loaded the hash since current->trap
> - * is not a 0x400 or 0x300, so just call hash_page_mm() here.
> - */
> - access = _PAGE_PRESENT | _PAGE_READ;
> - if (dsisr & CXL_PSL_DSISR_An_S)
> - access |= _PAGE_WRITE;
> -
> - access |= _PAGE_PRIVILEGED;
> - if ((!ctx->kernel) || (REGION_ID(dar) == USER_REGION_ID))
> - access &= ~_PAGE_PRIVILEGED;
> -
> - if (dsisr & DSISR_NOHPTE)
> - inv_flags |= HPTE_NOHPTE_UPDATE;
> -
> - local_irq_save(flags);
> - hash_page_mm(mm, dar, access, 0x300, inv_flags);
> - local_irq_restore(flags);
> -
> + if (!radix_enabled()) {
> + /*
> + * update_mmu_cache() will not have loaded the hash since current->trap
> + * is not a 0x400 or 0x300, so just call hash_page_mm() here.
> + */
> + access = _PAGE_PRESENT | _PAGE_READ;
> + if (dsisr & CXL_PSL_DSISR_An_S)
> + access |= _PAGE_WRITE;
> +
> + access |= _PAGE_PRIVILEGED;
> + if ((!ctx->kernel) || (REGION_ID(dar) == USER_REGION_ID))
> + access &= ~_PAGE_PRIVILEGED;
> +
> + if (dsisr & DSISR_NOHPTE)
> + inv_flags |= HPTE_NOHPTE_UPDATE;
> +
> + local_irq_save(flags);
> + hash_page_mm(mm, dar, access, 0x300, inv_flags);
> + local_irq_restore(flags);
> + }
> pr_devel("Page fault successfully handled for pe: %i!\n", ctx->pe);
> cxl_ops->ack_irq(ctx, CXL_PSL_TFC_An_R, 0);
> }
> @@ -239,6 +240,15 @@ void cxl_handle_fault(struct work_struct *fault_work)
> else
> WARN(1, "cxl_handle_fault has nothing to handle\n");
> }
> + if (cxl_is_psl9(ctx->afu)) {
> + if ((dsisr & CXL_PSL9_DSISR_An_CO_MASK) &
> + (CXL_PSL9_DSISR_An_PF_SLR | CXL_PSL9_DSISR_An_PF_RGC |
> + CXL_PSL9_DSISR_An_PF_RGP | CXL_PSL9_DSISR_An_PF_HRH |
> + CXL_PSL9_DSISR_An_PF_STEG))
> + cxl_handle_page_fault(ctx, mm, dsisr, dar);
> + else
> + WARN(1, "cxl_handle_fault has nothing to handle\n");
> + }
It would read better if we introduce a cxl_is_page_fault(ctx, dsisr)
function, which would handle the p8 and p9 differences.
>
> if (mm)
> mmput(mm);
> diff --git a/drivers/misc/cxl/guest.c b/drivers/misc/cxl/guest.c
> index 3ad7381..f58b4b6c 100644
> --- a/drivers/misc/cxl/guest.c
> +++ b/drivers/misc/cxl/guest.c
> @@ -551,13 +551,13 @@ static int attach_afu_directed(struct cxl_context *ctx, u64 wed, u64 amr)
> elem->common.tid = cpu_to_be32(0); /* Unused */
> elem->common.pid = cpu_to_be32(pid);
> elem->common.csrp = cpu_to_be64(0); /* disable */
> - elem->common.aurp0 = cpu_to_be64(0); /* disable */
> - elem->common.aurp1 = cpu_to_be64(0); /* disable */
> + elem->common.u.psl8.aurp0 = cpu_to_be64(0); /* disable */
> + elem->common.u.psl8.aurp1 = cpu_to_be64(0); /* disable */
>
> cxl_prefault(ctx, wed);
>
> - elem->common.sstp0 = cpu_to_be64(ctx->sstp0);
> - elem->common.sstp1 = cpu_to_be64(ctx->sstp1);
> + elem->common.u.psl8.sstp0 = cpu_to_be64(ctx->sstp0);
> + elem->common.u.psl8.sstp1 = cpu_to_be64(ctx->sstp1);
>
> /*
> * Ensure we have at least one interrupt allocated to take faults for
> diff --git a/drivers/misc/cxl/irq.c b/drivers/misc/cxl/irq.c
> index fa9f8a2..1eb5168 100644
> --- a/drivers/misc/cxl/irq.c
> +++ b/drivers/misc/cxl/irq.c
> @@ -34,6 +34,59 @@ static irqreturn_t schedule_cxl_fault(struct cxl_context *ctx, u64 dsisr, u64 da
> return IRQ_HANDLED;
> }
>
> +irqreturn_t cxl_irq_psl9(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info)
> +{
> + u64 dsisr, dar;
> +
> + dsisr = irq_info->dsisr;
> + dar = irq_info->dar;
> +
> + trace_cxl_psl9_irq(ctx, irq, dsisr, dar);
> +
> + pr_devel("CXL interrupt %i for afu pe: %i DSISR: %#llx DAR: %#llx\n", irq, ctx->pe, dsisr, dar);
> +
> + if (dsisr & CXL_PSL9_DSISR_An_TF) {
> + pr_devel("CXL interrupt: Scheduling translation fault"
> + " handling for later (pe: %i)\n", ctx->pe);
> + return schedule_cxl_fault(ctx, dsisr, dar);
> + }
> +
> + if (dsisr & CXL_PSL9_DSISR_An_PE)
> + return cxl_ops->handle_psl_slice_error(ctx, dsisr,
> + irq_info->errstat);
> + if (dsisr & CXL_PSL9_DSISR_An_AE) {
> + pr_devel("CXL interrupt: AFU Error 0x%016llx\n", irq_info->afu_err);
> +
> + if (ctx->pending_afu_err) {
> + /*
> + * This shouldn't happen - the PSL treats these errors
> + * as fatal and will have reset the AFU, so there's not
> + * much point buffering multiple AFU errors.
> + * OTOH if we DO ever see a storm of these come in it's
> + * probably best that we log them somewhere:
> + */
> + dev_err_ratelimited(&ctx->afu->dev, "CXL AFU Error "
> + "undelivered to pe %i: 0x%016llx\n",
> + ctx->pe, irq_info->afu_err);
> + } else {
> + spin_lock(&ctx->lock);
> + ctx->afu_err = irq_info->afu_err;
> + ctx->pending_afu_err = 1;
> + spin_unlock(&ctx->lock);
> +
> + wake_up_all(&ctx->wq);
> + }
> +
> + cxl_ops->ack_irq(ctx, CXL_PSL_TFC_An_A, 0);
> + return IRQ_HANDLED;
> + }
> + if (dsisr & CXL_PSL9_DSISR_An_OC)
> + pr_devel("CXL interrupt: OS Context Warning\n");
> +
> + WARN(1, "Unhandled CXL PSL IRQ\n");
> + return IRQ_HANDLED;
> +}
> +
> irqreturn_t cxl_irq_psl8(int irq, struct cxl_context *ctx, struct cxl_irq_info *irq_info)
> {
> u64 dsisr, dar;
> diff --git a/drivers/misc/cxl/native.c b/drivers/misc/cxl/native.c
> index 51aefcb..b4f7569 100644
> --- a/drivers/misc/cxl/native.c
> +++ b/drivers/misc/cxl/native.c
> @@ -175,6 +175,23 @@ int cxl_psl_purge(struct cxl_afu *afu)
> cpu_relax();
> }
> }
> + if (cxl_is_psl9(afu)) {
> + if (dsisr & CXL_PSL9_DSISR_An_TF) {
> + dar = cxl_p2n_read(afu, CXL_PSL_DAR_An);
> + dev_notice(&afu->dev, "PSL purge terminating "
> + "pending translation, "
> + "DSISR: 0x%016llx, DAR: 0x%016llx\n",
> + dsisr, dar);
> + cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_AE);
> + } else if (dsisr) {
> + dev_notice(&afu->dev, "PSL purge acknowledging "
> + "pending non-translation fault, "
> + "DSISR: 0x%016llx\n", dsisr);
> + cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_A);
> + } else {
> + cpu_relax();
> + }
> + }
Looks like we could refactor a bit better to avoid code duplication.
Only one flag read on the dsisr seems to change between p8 and p9.
> PSL_CNTL = cxl_p1n_read(afu, CXL_PSL_SCNTL_An);
> }
> end = local_clock();
> @@ -267,6 +284,36 @@ void cxl_release_spa(struct cxl_afu *afu)
> }
> }
>
> +int cxl_invalidate_all_psl9(struct cxl *adapter)
> +{
> + unsigned long timeout = jiffies + (HZ * CXL_TIMEOUT);
> + u64 ierat;
> +
> + /* do not invalidate ERAT entries when not reloading on PERST */
So this looks wrong, as the function doesn't always do what the name
implies. I think we should take out the test "only do it if not
reloading on perst" out, and only call this function when required, and
explain that it is debug only and no longer required by CAIA2.
> + if (adapter->perst_loads_image)
> + return 0;
> +
> + pr_devel("CXL adapter - invalidation of all ERAT entries\n");
> +
> + /* Invalidates all ERAT entries for Radix or HPT */
> + ierat = CXL_XSL9_IERAT_IALL;
> + if (radix_enabled())
> + ierat |= CXL_XSL9_IERAT_INVR;
> + cxl_p1_write(adapter, CXL_XSL9_IERAT, ierat);
> +
> + while (cxl_p1_read(adapter, CXL_XSL9_IERAT) & CXL_XSL9_IERAT_IINPROG) {
> + if (time_after_eq(jiffies, timeout)) {
> + dev_warn(&adapter->dev,
> + "WARNING: CXL adapter invalidation of all ERAT entries timed out!\n");
> + return -EBUSY;
> + }
> + if (!cxl_ops->link_ok(adapter, NULL))
> + return -EIO;
> + cpu_relax();
> + }
> + return 0;
> +}
> +
> int cxl_invalidate_all_psl8(struct cxl *adapter)
> {
> unsigned long timeout = jiffies + (HZ * CXL_TIMEOUT);
> @@ -553,10 +600,19 @@ static u64 calculate_sr(struct cxl_context *ctx)
> sr |= (mfmsr() & MSR_SF) | CXL_PSL_SR_An_HV;
> } else {
> sr |= CXL_PSL_SR_An_PR | CXL_PSL_SR_An_R;
> - sr &= ~(CXL_PSL_SR_An_HV);
> + if (radix_enabled())
> + sr |= CXL_PSL_SR_An_HV;
> + else
> + sr &= ~(CXL_PSL_SR_An_HV);
> if (!test_tsk_thread_flag(current, TIF_32BIT))
> sr |= CXL_PSL_SR_An_SF;
> }
> + if (cxl_is_psl9(ctx->afu)) {
> + if (radix_enabled())
> + sr |= CXL_PSL_SR_An_XLAT_ror;
> + else
> + sr |= CXL_PSL_SR_An_XLAT_hpt;
> + }
> return sr;
> }
>
> @@ -589,6 +645,70 @@ static void update_ivtes_directed(struct cxl_context *ctx)
> WARN_ON(add_process_element(ctx));
> }
>
> +static int process_element_entry(struct cxl_context *ctx, u64 wed, u64 amr)
> +{
> + u32 pid;
> +
> + cxl_assign_psn_space(ctx);
> +
> + ctx->elem->ctxtime = 0; /* disable */
> + ctx->elem->lpid = cpu_to_be32(mfspr(SPRN_LPID));
> + ctx->elem->haurp = 0; /* disable */
> +
> + if (ctx->kernel)
> + pid = 0;
> + else {
> + if (ctx->mm == NULL) {
> + pr_devel("%s: unable to get mm for pe=%d pid=%i\n",
> + __func__, ctx->pe, pid_nr(ctx->pid));
> + return -EINVAL;
> + }
> + pid = ctx->mm->context.id;
> + }
> +
> + ctx->elem->common.tid = 0;
> + ctx->elem->common.pid = cpu_to_be32(pid);
> +
> + ctx->elem->sr = cpu_to_be64(calculate_sr(ctx));
> +
> + ctx->elem->common.csrp = 0; /* disable */
> +
> + cxl_prefault(ctx, wed);
> +
> + /*
> + * Ensure we have the multiplexed PSL interrupt set up to take faults
> + * for kernel contexts that may not have allocated any AFU IRQs at all:
> + */
> + if (ctx->irqs.range[0] == 0) {
> + ctx->irqs.offset[0] = ctx->afu->native->psl_hwirq;
> + ctx->irqs.range[0] = 1;
> + }
> +
> + ctx->elem->common.amr = cpu_to_be64(amr);
> + ctx->elem->common.wed = cpu_to_be64(wed);
> +
> + return 0;
> +}
> +
> +int cxl_attach_afu_directed_psl9(struct cxl_context *ctx, u64 wed, u64 amr)
> +{
> + int result;
> +
> + /* fill the process element entry */
> + result = process_element_entry(ctx, wed, amr);
> + if (result)
> + return result;
> +
> + update_ivtes_directed(ctx);
> +
> + /* first guy needs to enable */
> + result = cxl_ops->afu_check_and_enable(ctx->afu);
> + if (result)
> + return result;
> +
> + return add_process_element(ctx);
> +}
> +
> int cxl_attach_afu_directed_psl8(struct cxl_context *ctx, u64 wed, u64 amr)
> {
> u32 pid;
> @@ -599,7 +719,7 @@ int cxl_attach_afu_directed_psl8(struct cxl_context *ctx, u64 wed, u64 amr)
> ctx->elem->ctxtime = 0; /* disable */
> ctx->elem->lpid = cpu_to_be32(mfspr(SPRN_LPID));
> ctx->elem->haurp = 0; /* disable */
> - ctx->elem->sdr = cpu_to_be64(mfspr(SPRN_SDR1));
> + ctx->elem->u.sdr = cpu_to_be64(mfspr(SPRN_SDR1));
>
> pid = current->pid;
> if (ctx->kernel)
> @@ -610,13 +730,13 @@ int cxl_attach_afu_directed_psl8(struct cxl_context *ctx, u64 wed, u64 amr)
> ctx->elem->sr = cpu_to_be64(calculate_sr(ctx));
>
> ctx->elem->common.csrp = 0; /* disable */
> - ctx->elem->common.aurp0 = 0; /* disable */
> - ctx->elem->common.aurp1 = 0; /* disable */
> + ctx->elem->common.u.psl8.aurp0 = 0; /* disable */
> + ctx->elem->common.u.psl8.aurp1 = 0; /* disable */
>
> cxl_prefault(ctx, wed);
>
> - ctx->elem->common.sstp0 = cpu_to_be64(ctx->sstp0);
> - ctx->elem->common.sstp1 = cpu_to_be64(ctx->sstp1);
> + ctx->elem->common.u.psl8.sstp0 = cpu_to_be64(ctx->sstp0);
> + ctx->elem->common.u.psl8.sstp1 = cpu_to_be64(ctx->sstp1);
>
> /*
> * Ensure we have the multiplexed PSL interrupt set up to take faults
> @@ -682,6 +802,32 @@ static int deactivate_afu_directed(struct cxl_afu *afu)
> return 0;
> }
>
> +int cxl_activate_dedicated_process_psl9(struct cxl_afu *afu)
> +{
> + dev_info(&afu->dev, "Activating dedicated process mode\n");
> +
> + /* If XSL is set to dedicated mode (Set in PSL_SCNTL reg), the
> + * XSL and AFU are programmed to work with a single context.
> + * The context information should be configured in the SPA area
> + * index 0 (so PSL_SPAP must be configured before enabling the
> + * AFU).
> + */
> + afu->num_procs = 1;
> + if (afu->native->spa == NULL) {
> + if (cxl_alloc_spa(afu))
> + return -ENOMEM;
> + }
> + attach_spa(afu);
> +
> + cxl_p1n_write(afu, CXL_PSL_SCNTL_An, CXL_PSL_SCNTL_An_PM_Process);
> + cxl_p1n_write(afu, CXL_PSL_ID_An, CXL_PSL_ID_An_F | CXL_PSL_ID_An_L);
> +
> + afu->current_mode = CXL_MODE_DEDICATED;
> + afu->num_procs = 1;
There are 2 occurences of "afu->num_procs = 1" (the first one is ~10
lines above). It needs to be defined before calling cxl_alloc_spa(), so
we should get rid of the 2nd occurence.
> +
> + return cxl_chardev_d_afu_add(afu);
> +}
> +
> int cxl_activate_dedicated_process_psl8(struct cxl_afu *afu)
> {
> dev_info(&afu->dev, "Activating dedicated process mode\n");
> @@ -705,6 +851,16 @@ int cxl_activate_dedicated_process_psl8(struct cxl_afu *afu)
> return cxl_chardev_d_afu_add(afu);
> }
>
> +void cxl_update_dedicated_ivtes_psl9(struct cxl_context *ctx)
> +{
> + int r;
> +
> + for (r = 0; r < CXL_IRQ_RANGES; r++) {
> + ctx->elem->ivte_offsets[r] = cpu_to_be16(ctx->irqs.offset[r]);
> + ctx->elem->ivte_ranges[r] = cpu_to_be16(ctx->irqs.range[r]);
> + }
> +}
> +
> void cxl_update_dedicated_ivtes_psl8(struct cxl_context *ctx)
> {
> struct cxl_afu *afu = ctx->afu;
> @@ -721,6 +877,26 @@ void cxl_update_dedicated_ivtes_psl8(struct cxl_context *ctx)
> ((u64)ctx->irqs.range[3] & 0xffff));
> }
>
> +int cxl_attach_dedicated_process_psl9(struct cxl_context *ctx, u64 wed, u64 amr)
> +{
> + struct cxl_afu *afu = ctx->afu;
> + int result;
> +
> + /* fill the process element entry */
> + result = process_element_entry(ctx, wed, amr);
> + if (result)
> + return result;
> +
> + if (ctx->afu->adapter->native->sl_ops->update_dedicated_ivtes)
> + afu->adapter->native->sl_ops->update_dedicated_ivtes(ctx);
> +
> + result = cxl_ops->afu_reset(afu);
> + if (result)
> + return result;
> +
> + return afu_enable(afu);
> +}
> +
> int cxl_attach_dedicated_process_psl8(struct cxl_context *ctx, u64 wed, u64 amr)
> {
> struct cxl_afu *afu = ctx->afu;
> @@ -892,6 +1068,21 @@ static int native_get_irq_info(struct cxl_afu *afu, struct cxl_irq_info *info)
> return 0;
> }
>
> +void cxl_native_irq_dump_regs_psl9(struct cxl_context *ctx)
> +{
> + u64 fir1, fir2, serr;
> +
> + fir1 = cxl_p1_read(ctx->afu->adapter, CXL_PSL9_FIR1);
> + fir2 = cxl_p1_read(ctx->afu->adapter, CXL_PSL9_FIR2);
> +
> + dev_crit(&ctx->afu->dev, "PSL_FIR1: 0x%016llx\n", fir1);
> + dev_crit(&ctx->afu->dev, "PSL_FIR2: 0x%016llx\n", fir2);
> + if (ctx->afu->adapter->native->sl_ops->register_serr_irq) {
> + serr = cxl_p1n_read(ctx->afu, CXL_PSL_SERR_An);
> + cxl_afu_decode_psl_serr(ctx->afu, serr);
> + }
> +}
> +
> void cxl_native_irq_dump_regs_psl8(struct cxl_context *ctx)
> {
> u64 fir1, fir2, fir_slice, serr, afu_debug;
> @@ -928,6 +1119,16 @@ static irqreturn_t native_handle_psl_slice_error(struct cxl_context *ctx,
> return cxl_ops->ack_irq(ctx, 0, errstat);
> }
>
> +irqreturn_t cxl_fail_irq_psl9(struct cxl_afu *afu, struct cxl_irq_info *irq_info)
> +{
> + if (irq_info->dsisr & CXL_PSL9_DSISR_An_TF)
> + cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_AE);
> + else
> + cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_A);
> +
> + return IRQ_HANDLED;
> +}
> +
Similarly to the first comment I made on native.c, the only difference
between the p8 and p9 version of this function is the flag read on the
dsisr. Couldn't we have a cxl_is_translation_fault(afu, dsisr) and merge
the 2 implementations (i.e. get rid of one callback)?
> irqreturn_t cxl_fail_irq_psl8(struct cxl_afu *afu, struct cxl_irq_info *irq_info)
> {
> if (irq_info->dsisr & CXL_PSL_DSISR_TRANS)
> @@ -999,6 +1200,9 @@ static void native_irq_wait(struct cxl_context *ctx)
> if (cxl_is_psl8(ctx->afu) &&
> ((dsisr & CXL_PSL_DSISR_PENDING) == 0))
> return;
> + if (cxl_is_psl9(ctx->afu) &&
> + ((dsisr & CXL_PSL9_DSISR_PENDING) == 0))
> + return;
> /*
> * We are waiting for the workqueue to process our
> * irq, so need to let that run here.
> @@ -1127,6 +1331,8 @@ int cxl_native_register_serr_irq(struct cxl_afu *afu)
> serr = cxl_p1n_read(afu, CXL_PSL_SERR_An);
> if (cxl_is_power8())
> serr = (serr & 0x00ffffffffff0000ULL) | (afu->serr_hwirq & 0xffff);
> + if (cxl_is_power9())
> + serr = (serr & ~0x0000000075010000ULL) | (afu->serr_hwirq & 0xffff);
This could use a comment, as it's not obvious why it's different on p8
(the reset value is not obvious).
Also, I don't think the p9 version if sufficient: we should explicitly
clear the bottom 16 bits before or'ing the interrupt number.
> cxl_p1n_write(afu, CXL_PSL_SERR_An, serr);
>
> return 0;
> diff --git a/drivers/misc/cxl/pci.c b/drivers/misc/cxl/pci.c
> index 5880092..c9f6215 100644
> --- a/drivers/misc/cxl/pci.c
> +++ b/drivers/misc/cxl/pci.c
> @@ -60,7 +60,7 @@
> #define CXL_VSEC_PROTOCOL_MASK 0xe0
> #define CXL_VSEC_PROTOCOL_1024TB 0x80
> #define CXL_VSEC_PROTOCOL_512TB 0x40
> -#define CXL_VSEC_PROTOCOL_256TB 0x20 /* Power 8 uses this */
> +#define CXL_VSEC_PROTOCOL_256TB 0x20 /* Power 8/9 uses this */
> #define CXL_VSEC_PROTOCOL_ENABLE 0x01
>
> #define CXL_READ_VSEC_PSL_REVISION(dev, vsec, dest) \
> @@ -326,14 +326,20 @@ static void dump_afu_descriptor(struct cxl_afu *afu)
>
> #define P8_CAPP_UNIT0_ID 0xBA
> #define P8_CAPP_UNIT1_ID 0XBE
> +#define P9_CAPP_UNIT0_ID 0xC0
> +#define P9_CAPP_UNIT1_ID 0xE0
>
> -static u64 get_capp_unit_id(struct device_node *np)
> +static u32 get_phb_index(struct device_node *np)
> {
> u32 phb_index;
>
> if (of_property_read_u32(np, "ibm,phb-index", &phb_index))
> - return 0;
> + return -ENODEV;
> + return phb_index;
> +}
>
> +static u64 get_capp_unit_id(struct device_node *np, u32 phb_index)
> +{
> /*
> * POWER 8:
> * - For chips other than POWER8NVL, we only have CAPP 0,
> @@ -352,10 +358,25 @@ static u64 get_capp_unit_id(struct device_node *np)
> return P8_CAPP_UNIT1_ID;
> }
>
> + /*
> + * POWER 9:
> + * PEC0 (PHB0). Capp ID = CAPP0 (0b1100_0000)
> + * PEC1 (PHB1 - PHB2). No capi mode
> + * PEC2 (PHB3 - PHB4 - PHB5): Capi mode on PHB3 only. Capp ID = CAPP1 (0b1110_0000)
> + */
> + if (cxl_is_power9()) {
> + if (phb_index == 0)
> + return P9_CAPP_UNIT0_ID;
> +
> + if (phb_index == 3)
> + return P9_CAPP_UNIT1_ID;
> + }
> +
> return 0;
> }
>
> -static int calc_capp_routing(struct pci_dev *dev, u64 *chipid, u64 *capp_unit_id)
> +static int calc_capp_routing(struct pci_dev *dev, u64 *chipid,
> + u32 *phb_index, u64 *capp_unit_id)
> {
> struct device_node *np;
> const __be32 *prop;
> @@ -367,8 +388,16 @@ static int calc_capp_routing(struct pci_dev *dev, u64 *chipid, u64 *capp_unit_id
> np = of_get_next_parent(np);
> if (!np)
> return -ENODEV;
> +
> *chipid = be32_to_cpup(prop);
> - *capp_unit_id = get_capp_unit_id(np);
> +
> + *phb_index = get_phb_index(np);
> + if (*phb_index == -ENODEV) {
> + pr_err("cxl: invalid phb index\n");
> + return -ENODEV;
> + }
> +
> + *capp_unit_id = get_capp_unit_id(np, *phb_index);
> of_node_put(np);
> if (!*capp_unit_id) {
> pr_err("cxl: invalid capp unit id\n");
> @@ -378,14 +407,99 @@ static int calc_capp_routing(struct pci_dev *dev, u64 *chipid, u64 *capp_unit_id
> return 0;
> }
>
> +static int init_implementation_adapter_regs_psl9(struct cxl *adapter, struct pci_dev *dev)
> +{
> + u64 xsl_dsnctl, psl_fircntl;
> + u64 chipid;
> + u32 phb_index;
> + u64 capp_unit_id;
> + int rc;
> +
> + rc = calc_capp_routing(dev, &chipid, &phb_index, &capp_unit_id);
> + if (rc)
> + return rc;
> +
> + /* CAPI Identifier bits [0:7]
> + * bit 61:60 MSI bits --> 0
> + * bit 59 TVT selector --> 0
> + */
> + /* Tell XSL where to route data to.
> + * The field chipid should match the PHB CAPI_CMPM register
> + */
> + xsl_dsnctl = ((u64)0x2 << (63-7)); /* Bit 57 */
> + xsl_dsnctl |= (capp_unit_id << (63-15));
> +
> + /* nMMU_ID Defaults to: b’000001001’*/
> + xsl_dsnctl |= ((u64)0x09 << (63-28));
> +
> + /* Used to identify CAPI packets which should be sorted into
> + * the Non-Blocking queues by the PHB. This field should match
> + * the PHB PBL_NBW_CMPM register
> + * nbwind=0x03, bits [57:58], must include capi indicator
> + */
> + xsl_dsnctl |= ((u64)0x03 << (63-47));
> +
> + /* Upper 16b address bits of ASB_Notify messages sent to the
> + * system. Need to match the PHB’s ASN Compare/Mask Register.
> + */
> + xsl_dsnctl |= ((u64)0x04 << (63-55));
> +
> + cxl_p1_write(adapter, CXL_XSL9_DSNCTL, xsl_dsnctl);
> +
> + /* Set fir_cntl to recommended value for production env */
> + psl_fircntl = (0x2ULL << (63-3)); /* ce_report */
> + psl_fircntl |= (0x1ULL << (63-6)); /* FIR_report */
> + psl_fircntl |= 0x1ULL; /* ce_thresh */
> + cxl_p1_write(adapter, CXL_PSL9_FIR_CNTL, psl_fircntl);
> +
> + /* vccredits=0x1 pcklat=0x4 */
> + cxl_p1_write(adapter, CXL_PSL9_DSNDCTL, 0x0000000000001810ULL);
> +
> + /* For debugging with trace arrays.
> + * Configure RX trace 0 segmented mode.
> + * Configure CT trace 0 segmented mode.
> + * Configure LA0 trace 0 segmented mode.
> + * Configure LA1 trace 0 segmented mode.
> + */
> + cxl_p1_write(adapter, CXL_PSL9_TRACECFG, 0x8040800080000000ULL);
> + cxl_p1_write(adapter, CXL_PSL9_TRACECFG, 0x8040800080000003ULL);
> + cxl_p1_write(adapter, CXL_PSL9_TRACECFG, 0x8040800080000005ULL);
> + cxl_p1_write(adapter, CXL_PSL9_TRACECFG, 0x8040800080000006ULL);
> +
> + /* A response to an ASB_Notify request is returned by the
> + * system as an MMIO write to the address defined in
> + * the PSL_TNR_ADDR register
> + */
> + /* PSL_TNR_ADDR */
> +
> + /* NORST */
> + cxl_p1_write(adapter, CXL_PSL9_DEBUG, 0x8000000000000000ULL);
> +
> + /* allocate the apc machines. For PHB0, let us keep the APC
> + * allocation setup as it is. For PHB3, we need to disable Rd
> + * machines
> + */
> + if (phb_index == 3) {
> + cxl_p1_write(adapter, CXL_PSL9_APCDEDALLOC, 0x8000808200000000ULL);
> + cxl_p1_write(adapter, CXL_PSL9_APCDEDTYPE, 0x40000003FFFF0000ULL);
> + }
> +
> + /* Disable vc dd1 fix */
> + if ((cxl_is_power9() && cpu_has_feature(CPU_FTR_POWER9_DD1)))
> + cxl_p1_write(adapter, CXL_PSL9_GP_CT, 0x0400000000000001ULL);
> +
> + return 0;
> +}
> +
> static int init_implementation_adapter_regs_psl8(struct cxl *adapter, struct pci_dev *dev)
> {
> u64 psl_dsnctl, psl_fircntl;
> u64 chipid;
> + u32 phb_index;
> u64 capp_unit_id;
> int rc;
>
> - rc = calc_capp_routing(dev, &chipid, &capp_unit_id);
> + rc = calc_capp_routing(dev, &chipid, &phb_index, &capp_unit_id);
> if (rc)
> return rc;
>
> @@ -414,10 +528,11 @@ static int init_implementation_adapter_regs_xsl(struct cxl *adapter, struct pci_
> {
> u64 xsl_dsnctl;
> u64 chipid;
> + u32 phb_index;
> u64 capp_unit_id;
> int rc;
>
> - rc = calc_capp_routing(dev, &chipid, &capp_unit_id);
> + rc = calc_capp_routing(dev, &chipid, &phb_index, &capp_unit_id);
> if (rc)
> return rc;
>
> @@ -435,6 +550,12 @@ static int init_implementation_adapter_regs_xsl(struct cxl *adapter, struct pci_
> /* For the PSL this is a multiple for 0 < n <= 7: */
> #define PSL_2048_250MHZ_CYCLES 1
>
> +static void write_timebase_ctrl_psl9(struct cxl *adapter)
> +{
> + cxl_p1_write(adapter, CXL_PSL9_TB_CTLSTAT,
> + TBSYNC_CNT(2 * PSL_2048_250MHZ_CYCLES));
> +}
> +
> static void write_timebase_ctrl_psl8(struct cxl *adapter)
> {
> cxl_p1_write(adapter, CXL_PSL_TB_CTLSTAT,
> @@ -456,6 +577,11 @@ static void write_timebase_ctrl_xsl(struct cxl *adapter)
> TBSYNC_CNT(XSL_4000_CLOCKS));
> }
>
> +static u64 timebase_read_psl9(struct cxl *adapter)
> +{
> + return cxl_p1_read(adapter, CXL_PSL9_Timebase);
> +}
> +
> static u64 timebase_read_psl8(struct cxl *adapter)
> {
> return cxl_p1_read(adapter, CXL_PSL_Timebase);
> @@ -514,6 +640,11 @@ static void cxl_setup_psl_timebase(struct cxl *adapter, struct pci_dev *dev)
> return;
> }
>
> +static int init_implementation_afu_regs_psl9(struct cxl_afu *afu)
> +{
> + return 0;
> +}
> +
> static int init_implementation_afu_regs_psl8(struct cxl_afu *afu)
> {
> /* read/write masks for this slice */
> @@ -612,7 +743,7 @@ static int setup_cxl_bars(struct pci_dev *dev)
> /*
> * BAR 4/5 has a special meaning for CXL and must be programmed with a
> * special value corresponding to the CXL protocol address range.
> - * For POWER 8 that means bits 48:49 must be set to 10
> + * For POWER 8/9 that means bits 48:49 must be set to 10
> */
> pci_write_config_dword(dev, PCI_BASE_ADDRESS_4, 0x00000000);
> pci_write_config_dword(dev, PCI_BASE_ADDRESS_5, 0x00020000);
> @@ -997,6 +1128,52 @@ static int cxl_afu_descriptor_looks_ok(struct cxl_afu *afu)
> return 0;
> }
>
> +static int sanitise_afu_regs_psl9(struct cxl_afu *afu)
> +{
> + u64 reg;
> +
> + /*
> + * Clear out any regs that contain either an IVTE or address or may be
> + * waiting on an acknowledgment to try to be a bit safer as we bring
> + * it online
> + */
> + reg = cxl_p2n_read(afu, CXL_AFU_Cntl_An);
> + if ((reg & CXL_AFU_Cntl_An_ES_MASK) != CXL_AFU_Cntl_An_ES_Disabled) {
> + dev_warn(&afu->dev, "WARNING: AFU was not disabled: %#016llx\n", reg);
> + if (cxl_ops->afu_reset(afu))
> + return -EIO;
> + if (cxl_afu_disable(afu))
> + return -EIO;
> + if (cxl_psl_purge(afu))
> + return -EIO;
> + }
> + cxl_p1n_write(afu, CXL_PSL_SPAP_An, 0x0000000000000000);
> + cxl_p1n_write(afu, CXL_PSL_AMBAR_An, 0x0000000000000000);
> + reg = cxl_p2n_read(afu, CXL_PSL_DSISR_An);
> + if (reg) {
> + dev_warn(&afu->dev, "AFU had pending DSISR: %#016llx\n", reg);
> + if (reg & CXL_PSL9_DSISR_An_TF)
> + cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_AE);
> + else
> + cxl_p2n_write(afu, CXL_PSL_TFC_An, CXL_PSL_TFC_An_A);
> + }
> + if (afu->adapter->native->sl_ops->register_serr_irq) {
> + reg = cxl_p1n_read(afu, CXL_PSL_SERR_An);
> + if (reg) {
> + if (reg & ~0x000000007501ffff)
We could make it easier if more error bits are added later by just
looking at the first 32 bits.
The rest of the patch looks good to me.
Fred
> + dev_warn(&afu->dev, "AFU had pending SERR: %#016llx\n", reg);
> + cxl_p1n_write(afu, CXL_PSL_SERR_An, reg & ~0xffff);
> + }
> + }
> + reg = cxl_p2n_read(afu, CXL_PSL_ErrStat_An);
> + if (reg) {
> + dev_warn(&afu->dev, "AFU had pending error status: %#016llx\n", reg);
> + cxl_p2n_write(afu, CXL_PSL_ErrStat_An, reg);
> + }
> +
> + return 0;
> +}
> +
> static int sanitise_afu_regs_psl8(struct cxl_afu *afu)
> {
> u64 reg;
> @@ -1393,6 +1570,9 @@ static bool cxl_compatible_caia_version(struct cxl *adapter)
> if (cxl_is_power8() && (adapter->caia_major == 1))
> return true;
>
> + if (cxl_is_power9() && (adapter->caia_major == 2))
> + return true;
> +
> return false;
> }
>
> @@ -1546,6 +1726,30 @@ static void cxl_deconfigure_adapter(struct cxl *adapter)
> pci_disable_device(pdev);
> }
>
> +static const struct cxl_service_layer_ops psl9_ops = {
> + .adapter_regs_init = init_implementation_adapter_regs_psl9,
> + .invalidate_all = cxl_invalidate_all_psl9,
> + .afu_regs_init = init_implementation_afu_regs_psl9,
> + .sanitise_afu_regs = sanitise_afu_regs_psl9,
> + .register_serr_irq = cxl_native_register_serr_irq,
> + .release_serr_irq = cxl_native_release_serr_irq,
> + .handle_interrupt = cxl_irq_psl9,
> + .fail_irq = cxl_fail_irq_psl9,
> + .activate_dedicated_process = cxl_activate_dedicated_process_psl9,
> + .attach_afu_directed = cxl_attach_afu_directed_psl9,
> + .attach_dedicated_process = cxl_attach_dedicated_process_psl9,
> + .update_dedicated_ivtes = cxl_update_dedicated_ivtes_psl9,
> + .debugfs_add_adapter_regs = cxl_debugfs_add_adapter_regs_psl9,
> + .debugfs_add_afu_regs = cxl_debugfs_add_afu_regs_psl9,
> + .psl_irq_dump_registers = cxl_native_irq_dump_regs_psl9,
> + .err_irq_dump_registers = cxl_native_err_irq_dump_regs,
> + .debugfs_stop_trace = cxl_stop_trace_psl9,
> + .write_timebase_ctrl = write_timebase_ctrl_psl9,
> + .timebase_read = timebase_read_psl9,
> + .capi_mode = OPAL_PHB_CAPI_MODE_CAPI,
> + .needs_reset_before_disable = true,
> +};
> +
> static const struct cxl_service_layer_ops psl8_ops = {
> .adapter_regs_init = init_implementation_adapter_regs_psl8,
> .invalidate_all = cxl_invalidate_all_psl8,
> @@ -1589,6 +1793,9 @@ static void set_sl_ops(struct cxl *adapter, struct pci_dev *dev)
> if (cxl_is_power8()) {
> dev_info(&dev->dev, "Device uses a PSL8\n");
> adapter->native->sl_ops = &psl8_ops;
> + } else {
> + dev_info(&dev->dev, "Device uses a PSL9\n");
> + adapter->native->sl_ops = &psl9_ops;
> }
> }
> }
> @@ -1744,6 +1951,11 @@ static int cxl_probe(struct pci_dev *dev, const struct pci_device_id *id)
> return -ENODEV;
> }
>
> + if (cxl_is_power9() && !radix_enabled()) {
> + dev_info(&dev->dev, "Only Radix mode supported\n");
> + return -ENODEV;
> + }
> +
> if (cxl_verbose)
> dump_cxl_config_space(dev);
>
> diff --git a/drivers/misc/cxl/trace.h b/drivers/misc/cxl/trace.h
> index 751d611..b8e300a 100644
> --- a/drivers/misc/cxl/trace.h
> +++ b/drivers/misc/cxl/trace.h
> @@ -17,6 +17,15 @@
>
> #include "cxl.h"
>
> +#define dsisr_psl9_flags(flags) \
> + __print_flags(flags, "|", \
> + { CXL_PSL9_DSISR_An_CO_MASK, "FR" }, \
> + { CXL_PSL9_DSISR_An_TF, "TF" }, \
> + { CXL_PSL9_DSISR_An_PE, "PE" }, \
> + { CXL_PSL9_DSISR_An_AE, "AE" }, \
> + { CXL_PSL9_DSISR_An_OC, "OC" }, \
> + { CXL_PSL9_DSISR_An_S, "S" })
> +
> #define DSISR_FLAGS \
> { CXL_PSL_DSISR_An_DS, "DS" }, \
> { CXL_PSL_DSISR_An_DM, "DM" }, \
> @@ -154,6 +163,40 @@ TRACE_EVENT(cxl_afu_irq,
> )
> );
>
> +TRACE_EVENT(cxl_psl9_irq,
> + TP_PROTO(struct cxl_context *ctx, int irq, u64 dsisr, u64 dar),
> +
> + TP_ARGS(ctx, irq, dsisr, dar),
> +
> + TP_STRUCT__entry(
> + __field(u8, card)
> + __field(u8, afu)
> + __field(u16, pe)
> + __field(int, irq)
> + __field(u64, dsisr)
> + __field(u64, dar)
> + ),
> +
> + TP_fast_assign(
> + __entry->card = ctx->afu->adapter->adapter_num;
> + __entry->afu = ctx->afu->slice;
> + __entry->pe = ctx->pe;
> + __entry->irq = irq;
> + __entry->dsisr = dsisr;
> + __entry->dar = dar;
> + ),
> +
> + TP_printk("afu%i.%i pe=%i irq=%i dsisr=0x%016llx dsisr=%s dar=0x%016llx",
> + __entry->card,
> + __entry->afu,
> + __entry->pe,
> + __entry->irq,
> + __entry->dsisr,
> + dsisr_psl9_flags(__entry->dsisr),
> + __entry->dar
> + )
> +);
> +
> TRACE_EVENT(cxl_psl_irq,
> TP_PROTO(struct cxl_context *ctx, int irq, u64 dsisr, u64 dar),
>
^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH V2 0/6] cxl: Add support for Coherent Accelerator Interface Architecture 2.0
2017-03-14 11:08 [PATCH V2 0/6] cxl: Add support for Coherent Accelerator Interface Architecture 2.0 Christophe Lombard
` (5 preceding siblings ...)
2017-03-14 11:08 ` [PATCH V2 6/6] cxl: Add psl9 specific code Christophe Lombard
@ 2017-03-21 2:47 ` Andrew Donnellan
2017-03-21 9:47 ` christophe lombard
6 siblings, 1 reply; 20+ messages in thread
From: Andrew Donnellan @ 2017-03-21 2:47 UTC (permalink / raw)
To: Christophe Lombard, linuxppc-dev, fbarrat, imunsie
On 14/03/17 22:08, Christophe Lombard wrote:
> The first 3 patches are mostly cleanup and fixes, separating the
> psl8-specific code from the code which will also be used for psl9.
> Patches 4 restructure existing code, to easily add the psl
> implementation.
> Patch 5 and 6 rename and isolate implementation-specific code.
> Patch 7 introduces the core of the PSL9-specific code.
Is patch 1 (cxl: Read vsec perst load image) from the last version of
the series missing in this one?
--
Andrew Donnellan OzLabs, ADL Canberra
andrew.donnellan@au1.ibm.com IBM Australia Limited
^ permalink raw reply [flat|nested] 20+ messages in thread* Re: [PATCH V2 0/6] cxl: Add support for Coherent Accelerator Interface Architecture 2.0
2017-03-21 2:47 ` [PATCH V2 0/6] cxl: Add support for Coherent Accelerator Interface Architecture 2.0 Andrew Donnellan
@ 2017-03-21 9:47 ` christophe lombard
0 siblings, 0 replies; 20+ messages in thread
From: christophe lombard @ 2017-03-21 9:47 UTC (permalink / raw)
To: Andrew Donnellan, linuxppc-dev, fbarrat, imunsie
Le 21/03/2017 à 03:47, Andrew Donnellan a écrit :
> On 14/03/17 22:08, Christophe Lombard wrote:
>> The first 3 patches are mostly cleanup and fixes, separating the
>> psl8-specific code from the code which will also be used for psl9.
>> Patches 4 restructure existing code, to easily add the psl
>> implementation.
>> Patch 5 and 6 rename and isolate implementation-specific code.
>> Patch 7 introduces the core of the PSL9-specific code.
>
> Is patch 1 (cxl: Read vsec perst load image) from the last version of
> the series missing in this one?
>
>
Hum. Good point !. It will be present again in V3.
sorry about that.
^ permalink raw reply [flat|nested] 20+ messages in thread