2018-04-18 01:13:50

by Alastair D'Silva

[permalink] [raw]

Subject: [PATCH v2 3/7] powerpc: use task_pid_nr() for TID allocation

From: Alastair D'Silva <[email protected]>

The current implementation of TID allocation, using a global IDR, may
result in an errant process starving the system of available TIDs.
Instead, use task_pid_nr(), as mentioned by the original author. The
scenario described which prevented it's use is not applicable, as
set_thread_tidr can only be called after the task struct has been
populated.

Signed-off-by: Alastair D'Silva <[email protected]>
---
arch/powerpc/include/asm/switch_to.h | 1 -
arch/powerpc/kernel/process.c | 97 +-----------------------------------
2 files changed, 1 insertion(+), 97 deletions(-)

diff --git a/arch/powerpc/include/asm/switch_to.h b/arch/powerpc/include/asm/switch_to.h
index be8c9fa23983..5b03d8a82409 100644
--- a/arch/powerpc/include/asm/switch_to.h
+++ b/arch/powerpc/include/asm/switch_to.h
@@ -94,6 +94,5 @@ static inline void clear_task_ebb(struct task_struct *t)
extern int set_thread_uses_vas(void);

extern int set_thread_tidr(struct task_struct *t);
-extern void clear_thread_tidr(struct task_struct *t);

#endif /* _ASM_POWERPC_SWITCH_TO_H */
diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c
index 3b00da47699b..87f047fd2762 100644
--- a/arch/powerpc/kernel/process.c
+++ b/arch/powerpc/kernel/process.c
@@ -1496,103 +1496,12 @@ int set_thread_uses_vas(void)
}

#ifdef CONFIG_PPC64
-static DEFINE_SPINLOCK(vas_thread_id_lock);
-static DEFINE_IDA(vas_thread_ida);
-
-/*
- * We need to assign a unique thread id to each thread in a process.
- *
- * This thread id, referred to as TIDR, and separate from the Linux's tgid,
- * is intended to be used to direct an ASB_Notify from the hardware to the
- * thread, when a suitable event occurs in the system.
- *
- * One such event is a "paste" instruction in the context of Fast Thread
- * Wakeup (aka Core-to-core wake up in the Virtual Accelerator Switchboard
- * (VAS) in POWER9.
- *
- * To get a unique TIDR per process we could simply reuse task_pid_nr() but
- * the problem is that task_pid_nr() is not yet available copy_thread() is
- * called. Fixing that would require changing more intrusive arch-neutral
- * code in code path in copy_process()?.
- *
- * Further, to assign unique TIDRs within each process, we need an atomic
- * field (or an IDR) in task_struct, which again intrudes into the arch-
- * neutral code. So try to assign globally unique TIDRs for now.
- *
- * NOTE: TIDR 0 indicates that the thread does not need a TIDR value.
- * For now, only threads that expect to be notified by the VAS
- * hardware need a TIDR value and we assign values > 0 for those.
- */
-#define MAX_THREAD_CONTEXT ((1 << 16) - 1)
-static int assign_thread_tidr(void)
-{
- int index;
- int err;
- unsigned long flags;
-
-again:
- if (!ida_pre_get(&vas_thread_ida, GFP_KERNEL))
- return -ENOMEM;
-
- spin_lock_irqsave(&vas_thread_id_lock, flags);
- err = ida_get_new_above(&vas_thread_ida, 1, &index);
- spin_unlock_irqrestore(&vas_thread_id_lock, flags);
-
- if (err == -EAGAIN)
- goto again;
- else if (err)
- return err;
-
- if (index > MAX_THREAD_CONTEXT) {
- spin_lock_irqsave(&vas_thread_id_lock, flags);
- ida_remove(&vas_thread_ida, index);
- spin_unlock_irqrestore(&vas_thread_id_lock, flags);
- return -ENOMEM;
- }
-
- return index;
-}
-
-static void free_thread_tidr(int id)
-{
- unsigned long flags;
-
- spin_lock_irqsave(&vas_thread_id_lock, flags);
- ida_remove(&vas_thread_ida, id);
- spin_unlock_irqrestore(&vas_thread_id_lock, flags);
-}
-
-/*
- * Clear any TIDR value assigned to this thread.
- */
-void clear_thread_tidr(struct task_struct *t)
-{
- if (!t->thread.tidr)
- return;
-
- if (!cpu_has_feature(CPU_FTR_P9_TIDR)) {
- WARN_ON_ONCE(1);
- return;
- }
-
- mtspr(SPRN_TIDR, 0);
- free_thread_tidr(t->thread.tidr);
- t->thread.tidr = 0;
-}
-
-void arch_release_task_struct(struct task_struct *t)
-{
- clear_thread_tidr(t);
-}
-
/*
* Assign a unique TIDR (thread id) for task @t and set it in the thread
* structure. For now, we only support setting TIDR for 'current' task.
*/
int set_thread_tidr(struct task_struct *t)
{
- int rc;
-
if (!cpu_has_feature(CPU_FTR_P9_TIDR))
return -EINVAL;

@@ -1602,11 +1511,7 @@ int set_thread_tidr(struct task_struct *t)
if (t->thread.tidr)
return 0;

- rc = assign_thread_tidr();
- if (rc < 0)
- return rc;
-
- t->thread.tidr = rc;
+ t->thread.tidr = (u16)task_pid_nr(t);
mtspr(SPRN_TIDR, t->thread.tidr);

return 0;
--
2.14.3

2018-04-18 01:14:18

by Alastair D'Silva

[permalink] [raw]

2018-05-09 00:47:38

by Alastair D'Silva

[permalink] [raw]

Subject: [PATCH v3 7/7] ocxl: Document new OCXL IOCTLs

From: Alastair D'Silva <[email protected]>

Signed-off-by: Alastair D'Silva <[email protected]>
---
Documentation/accelerators/ocxl.rst | 11 +++++++++++
1 file changed, 11 insertions(+)

diff --git a/Documentation/accelerators/ocxl.rst b/Documentation/accelerators/ocxl.rst
index ddcc58d01cfb..14cefc020e2d 100644
--- a/Documentation/accelerators/ocxl.rst
+++ b/Documentation/accelerators/ocxl.rst
@@ -157,6 +157,17 @@ OCXL_IOCTL_GET_METADATA:
Obtains configuration information from the card, such at the size of
MMIO areas, the AFU version, and the PASID for the current context.

+OCXL_IOCTL_ENABLE_P9_WAIT:
+
+ Allows the AFU to wake a userspace thread executing 'wait'. Returns
+ information to userspace to allow it to configure the AFU. Note that
+ this is only available on POWER9.
+
+OCXL_IOCTL_GET_FEATURES:
+
+ Reports on which CPU features that affect OpenCAPI are usable from
+ userspace.
+

mmap
----
--
2.14.3

2018-05-09 00:47:58

by Alastair D'Silva

[permalink] [raw]

Subject: [PATCH v3 5/7] ocxl: Expose the thread_id needed for wait on POWER9

From: Alastair D'Silva <[email protected]>

In order to successfully issue as_notify, an AFU needs to know the TID
to notify, which in turn means that this information should be
available in userspace so it can be communicated to the AFU.

Signed-off-by: Alastair D'Silva <[email protected]>
---
drivers/misc/ocxl/context.c | 5 +++-
drivers/misc/ocxl/file.c | 53 +++++++++++++++++++++++++++++++++++++++
drivers/misc/ocxl/link.c | 36 ++++++++++++++++++++++++++
drivers/misc/ocxl/ocxl_internal.h | 1 +
include/misc/ocxl.h | 9 +++++++
include/uapi/misc/ocxl.h | 10 ++++++++
6 files changed, 113 insertions(+), 1 deletion(-)

diff --git a/drivers/misc/ocxl/context.c b/drivers/misc/ocxl/context.c
index 909e8807824a..95f74623113e 100644
--- a/drivers/misc/ocxl/context.c
+++ b/drivers/misc/ocxl/context.c
@@ -34,6 +34,8 @@ int ocxl_context_init(struct ocxl_context *ctx, struct ocxl_afu *afu,
mutex_init(&ctx->xsl_error_lock);
mutex_init(&ctx->irq_lock);
idr_init(&ctx->irq_idr);
+ ctx->tidr = 0;
+
/*
* Keep a reference on the AFU to make sure it's valid for the
* duration of the life of the context
@@ -65,6 +67,7 @@ int ocxl_context_attach(struct ocxl_context *ctx, u64 amr)
{
int rc;

+ // Locks both status & tidr
mutex_lock(&ctx->status_mutex);
if (ctx->status != OPENED) {
rc = -EIO;
@@ -72,7 +75,7 @@ int ocxl_context_attach(struct ocxl_context *ctx, u64 amr)
}

rc = ocxl_link_add_pe(ctx->afu->fn->link, ctx->pasid,
- current->mm->context.id, 0, amr, current->mm,
+ current->mm->context.id, ctx->tidr, amr, current->mm,
xsl_fault_error, ctx);
if (rc)
goto out;
diff --git a/drivers/misc/ocxl/file.c b/drivers/misc/ocxl/file.c
index 038509e5d031..eb409a469f21 100644
--- a/drivers/misc/ocxl/file.c
+++ b/drivers/misc/ocxl/file.c
@@ -5,6 +5,8 @@
#include <linux/sched/signal.h>
#include <linux/uaccess.h>
#include <uapi/misc/ocxl.h>
+#include <asm/reg.h>
+#include <asm/switch_to.h>
#include "ocxl_internal.h"

@@ -123,11 +125,55 @@ static long afu_ioctl_get_metadata(struct ocxl_context *ctx,
return 0;
}

+#ifdef CONFIG_PPC64
+static long afu_ioctl_enable_p9_wait(struct ocxl_context *ctx,
+ struct ocxl_ioctl_p9_wait __user *uarg)
+{
+ struct ocxl_ioctl_p9_wait arg;
+
+ memset(&arg, 0, sizeof(arg));
+
+ if (cpu_has_feature(CPU_FTR_P9_TIDR)) {
+ enum ocxl_context_status status;
+
+ // Locks both status & tidr
+ mutex_lock(&ctx->status_mutex);
+ if (!ctx->tidr) {
+ if (set_thread_tidr(current))
+ return -ENOENT;
+
+ ctx->tidr = current->thread.tidr;
+ }
+
+ status = ctx->status;
+ mutex_unlock(&ctx->status_mutex);
+
+ if (status == ATTACHED) {
+ int rc;
+ struct link *link = ctx->afu->fn->link;
+
+ rc = ocxl_link_update_pe(link, ctx->pasid, ctx->tidr);
+ if (rc)
+ return rc;
+ }
+
+ arg.thread_id = ctx->tidr;
+ } else
+ return -ENOENT;
+
+ if (copy_to_user(uarg, &arg, sizeof(arg)))
+ return -EFAULT;
+
+ return 0;
+}
+#endif
+
#define CMD_STR(x) (x == OCXL_IOCTL_ATTACH ? "ATTACH" : \
x == OCXL_IOCTL_IRQ_ALLOC ? "IRQ_ALLOC" : \
x == OCXL_IOCTL_IRQ_FREE ? "IRQ_FREE" : \
x == OCXL_IOCTL_IRQ_SET_FD ? "IRQ_SET_FD" : \
x == OCXL_IOCTL_GET_METADATA ? "GET_METADATA" : \
+ x == OCXL_IOCTL_ENABLE_P9_WAIT ? "ENABLE_P9_WAIT" : \
"UNKNOWN")

static long afu_ioctl(struct file *file, unsigned int cmd,
@@ -186,6 +232,13 @@ static long afu_ioctl(struct file *file, unsigned int cmd,
(struct ocxl_ioctl_metadata __user *) args);
break;

+#ifdef CONFIG_PPC64
+ case OCXL_IOCTL_ENABLE_P9_WAIT:
+ rc = afu_ioctl_enable_p9_wait(ctx,
+ (struct ocxl_ioctl_p9_wait __user *) args);
+ break;
+#endif
+
default:
rc = -EINVAL;
}
diff --git a/drivers/misc/ocxl/link.c b/drivers/misc/ocxl/link.c
index 656e8610eec2..88876ae8f330 100644
--- a/drivers/misc/ocxl/link.c
+++ b/drivers/misc/ocxl/link.c
@@ -544,6 +544,42 @@ int ocxl_link_add_pe(void *link_handle, int pasid, u32 pidr, u32 tidr,
}
EXPORT_SYMBOL_GPL(ocxl_link_add_pe);

+int ocxl_link_update_pe(void *link_handle, int pasid, __u16 tid)
+{
+ struct link *link = (struct link *) link_handle;
+ struct spa *spa = link->spa;
+ struct ocxl_process_element *pe;
+ int pe_handle, rc;
+
+ if (pasid > SPA_PASID_MAX)
+ return -EINVAL;
+
+ pe_handle = pasid & SPA_PE_MASK;
+ pe = spa->spa_mem + pe_handle;
+
+ mutex_lock(&spa->spa_lock);
+
+ pe->tid = tid;
+
+ /*
+ * The barrier makes sure the PE is updated
+ * before we clear the NPU context cache below, so that the
+ * old PE cannot be reloaded erroneously.
+ */
+ mb();
+
+ /*
+ * hook to platform code
+ * On powerpc, the entry needs to be cleared from the context
+ * cache of the NPU.
+ */
+ rc = pnv_ocxl_spa_remove_pe_from_cache(link->platform_data, pe_handle);
+ WARN_ON(rc);
+
+ mutex_unlock(&spa->spa_lock);
+ return rc;
+}
+
int ocxl_link_remove_pe(void *link_handle, int pasid)
{
struct link *link = (struct link *) link_handle;
diff --git a/drivers/misc/ocxl/ocxl_internal.h b/drivers/misc/ocxl/ocxl_internal.h
index 5d421824afd9..a32f2151029f 100644
--- a/drivers/misc/ocxl/ocxl_internal.h
+++ b/drivers/misc/ocxl/ocxl_internal.h
@@ -77,6 +77,7 @@ struct ocxl_context {
struct ocxl_xsl_error xsl_error;
struct mutex irq_lock;
struct idr irq_idr;
+ u16 tidr; // Thread ID used for P9 wait implementation
};

struct ocxl_process_element {
diff --git a/include/misc/ocxl.h b/include/misc/ocxl.h
index 51ccf76db293..9ff6ddc28e22 100644
--- a/include/misc/ocxl.h
+++ b/include/misc/ocxl.h
@@ -188,6 +188,15 @@ extern int ocxl_link_add_pe(void *link_handle, int pasid, u32 pidr, u32 tidr,
void (*xsl_err_cb)(void *data, u64 addr, u64 dsisr),
void *xsl_err_data);

+/**
+ * Update values within a Process Element
+ *
+ * link_handle: the link handle associated with the process element
+ * pasid: the PASID for the AFU context
+ * tid: the new thread id for the process element
+ */
+extern int ocxl_link_update_pe(void *link_handle, int pasid, __u16 tid);
+
/*
* Remove a Process Element from the Shared Process Area for a link
*/
diff --git a/include/uapi/misc/ocxl.h b/include/uapi/misc/ocxl.h
index 0af83d80fb3e..8d2748e69c84 100644
--- a/include/uapi/misc/ocxl.h
+++ b/include/uapi/misc/ocxl.h
@@ -48,6 +48,15 @@ struct ocxl_ioctl_metadata {
__u64 reserved[13]; // Total of 16*u64
};

+struct ocxl_ioctl_p9_wait {
+ __u16 thread_id; // The thread ID required to wake this thread
+ __u16 reserved1;
+ __u32 reserved2;
+ __u64 reserved3[3];
+};
+
+};
+
struct ocxl_ioctl_irq_fd {
__u64 irq_offset;
__s32 eventfd;
@@ -62,5 +71,6 @@ struct ocxl_ioctl_irq_fd {
#define OCXL_IOCTL_IRQ_FREE _IOW(OCXL_MAGIC, 0x12, __u64)
#define OCXL_IOCTL_IRQ_SET_FD _IOW(OCXL_MAGIC, 0x13, struct ocxl_ioctl_irq_fd)
#define OCXL_IOCTL_GET_METADATA _IOR(OCXL_MAGIC, 0x14, struct ocxl_ioctl_metadata)
+#define OCXL_IOCTL_ENABLE_P9_WAIT _IOR(OCXL_MAGIC, 0x15, struct ocxl_ioctl_p9_wait)

#endif /* _UAPI_MISC_OCXL_H */
--
2.14.3

2018-05-09 05:36:56

by Alastair D'Silva

[permalink] [raw]