p71924506
/
linux_old1
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
linux_old1/drivers/misc/cxl/guest.c

1203 lines
27 KiB
C
Raw Normal View History

cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
/*
* Copyright 2015 IBM Corp.
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version
* 2 of the License, or (at your option) any later version.
*/
#include <linux/spinlock.h>
#include <linux/uaccess.h>
#include <linux/delay.h>
#include "cxl.h"
#include "hcalls.h"
#include "trace.h"
cxl: Adapter failure handling Check the AFU state whenever an API is called. The hypervisor may issue a reset of the adapter when it detects a fault. When it happens, it launches an error recovery which will either move the AFU to a permanent failure state, or in the disabled state. If the AFU is found to be disabled, detach all existing contexts from it before issuing a AFU reset to re-enable it. Before detaching contexts, notify any kernel driver through the EEH callbacks of the AFU pci device. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:41 +08:00
#define CXL_ERROR_DETECTED_EVENT 1
#define CXL_SLOT_RESET_EVENT 2
#define CXL_RESUME_EVENT 3
static void pci_error_handlers(struct cxl_afu *afu,
int bus_error_event,
pci_channel_state_t state)
{
struct pci_dev *afu_dev;
if (afu->phb == NULL)
return;
list_for_each_entry(afu_dev, &afu->phb->bus->devices, bus_list) {
if (!afu_dev->driver)
continue;
switch (bus_error_event) {
case CXL_ERROR_DETECTED_EVENT:
afu_dev->error_state = state;
if (afu_dev->driver->err_handler &&
afu_dev->driver->err_handler->error_detected)
afu_dev->driver->err_handler->error_detected(afu_dev, state);
break;
case CXL_SLOT_RESET_EVENT:
afu_dev->error_state = state;
if (afu_dev->driver->err_handler &&
afu_dev->driver->err_handler->slot_reset)
afu_dev->driver->err_handler->slot_reset(afu_dev);
break;
case CXL_RESUME_EVENT:
if (afu_dev->driver->err_handler &&
afu_dev->driver->err_handler->resume)
afu_dev->driver->err_handler->resume(afu_dev);
break;
}
}
}
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
static irqreturn_t guest_handle_psl_slice_error(struct cxl_context *ctx, u64 dsisr,
u64 errstat)
{
pr_devel("in %s\n", __func__);
dev_crit(&ctx->afu->dev, "PSL ERROR STATUS: 0x%.16llx\n", errstat);
return cxl_ops->ack_irq(ctx, 0, errstat);
}
static ssize_t guest_collect_vpd(struct cxl *adapter, struct cxl_afu *afu,
void *buf, size_t len)
{
unsigned int entries, mod;
unsigned long **vpd_buf = NULL;
struct sg_list *le;
int rc = 0, i, tocopy;
u64 out = 0;
if (buf == NULL)
return -EINVAL;
/* number of entries in the list */
entries = len / SG_BUFFER_SIZE;
mod = len % SG_BUFFER_SIZE;
if (mod)
entries++;
if (entries > SG_MAX_ENTRIES) {
entries = SG_MAX_ENTRIES;
len = SG_MAX_ENTRIES * SG_BUFFER_SIZE;
mod = 0;
}
treewide: kzalloc() -> kcalloc() The kzalloc() function has a 2-factor argument form, kcalloc(). This patch replaces cases of: kzalloc(a * b, gfp) with: kcalloc(a * b, gfp) as well as handling cases of: kzalloc(a * b * c, gfp) with: kzalloc(array3_size(a, b, c), gfp) as it's slightly less ugly than: kzalloc_array(array_size(a, b), c, gfp) This does, however, attempt to ignore constant size factors like: kzalloc(4 * 1024, gfp) though any constants defined via macros get caught up in the conversion. Any factors with a sizeof() of "unsigned char", "char", and "u8" were dropped, since they're redundant. The Coccinelle script used for this was: // Fix redundant parens around sizeof(). @@ type TYPE; expression THING, E; @@ ( kzalloc( - (sizeof(TYPE)) * E + sizeof(TYPE) * E , ...) | kzalloc( - (sizeof(THING)) * E + sizeof(THING) * E , ...) ) // Drop single-byte sizes and redundant parens. @@ expression COUNT; typedef u8; typedef __u8; @@ ( kzalloc( - sizeof(u8) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(__u8) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(char) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(unsigned char) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(u8) * COUNT + COUNT , ...) | kzalloc( - sizeof(__u8) * COUNT + COUNT , ...) | kzalloc( - sizeof(char) * COUNT + COUNT , ...) | kzalloc( - sizeof(unsigned char) * COUNT + COUNT , ...) ) // 2-factor product with sizeof(type/expression) and identifier or constant. @@ type TYPE; expression THING; identifier COUNT_ID; constant COUNT_CONST; @@ ( - kzalloc + kcalloc ( - sizeof(TYPE) * (COUNT_ID) + COUNT_ID, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * COUNT_ID + COUNT_ID, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * (COUNT_CONST) + COUNT_CONST, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * COUNT_CONST + COUNT_CONST, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * (COUNT_ID) + COUNT_ID, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * COUNT_ID + COUNT_ID, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * (COUNT_CONST) + COUNT_CONST, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * COUNT_CONST + COUNT_CONST, sizeof(THING) , ...) ) // 2-factor product, only identifiers. @@ identifier SIZE, COUNT; @@ - kzalloc + kcalloc ( - SIZE * COUNT + COUNT, SIZE , ...) // 3-factor product with 1 sizeof(type) or sizeof(expression), with // redundant parens removed. @@ expression THING; identifier STRIDE, COUNT; type TYPE; @@ ( kzalloc( - sizeof(TYPE) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(TYPE) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(TYPE) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(TYPE) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(THING) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kzalloc( - sizeof(THING) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kzalloc( - sizeof(THING) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kzalloc( - sizeof(THING) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) ) // 3-factor product with 2 sizeof(variable), with redundant parens removed. @@ expression THING1, THING2; identifier COUNT; type TYPE1, TYPE2; @@ ( kzalloc( - sizeof(TYPE1) * sizeof(TYPE2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | kzalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | kzalloc( - sizeof(THING1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | kzalloc( - sizeof(THING1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | kzalloc( - sizeof(TYPE1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) | kzalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) ) // 3-factor product, only identifiers, with redundant parens removed. @@ identifier STRIDE, SIZE, COUNT; @@ ( kzalloc( - (COUNT) * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - (COUNT) * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - (COUNT) * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - (COUNT) * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) ) // Any remaining multi-factor products, first at least 3-factor products, // when they're not all constants... @@ expression E1, E2, E3; constant C1, C2, C3; @@ ( kzalloc(C1 * C2 * C3, ...) | kzalloc( - (E1) * E2 * E3 + array3_size(E1, E2, E3) , ...) | kzalloc( - (E1) * (E2) * E3 + array3_size(E1, E2, E3) , ...) | kzalloc( - (E1) * (E2) * (E3) + array3_size(E1, E2, E3) , ...) | kzalloc( - E1 * E2 * E3 + array3_size(E1, E2, E3) , ...) ) // And then all remaining 2 factors products when they're not all constants, // keeping sizeof() as the second factor argument. @@ expression THING, E1, E2; type TYPE; constant C1, C2, C3; @@ ( kzalloc(sizeof(THING) * C2, ...) | kzalloc(sizeof(TYPE) * C2, ...) | kzalloc(C1 * C2 * C3, ...) | kzalloc(C1 * C2, ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * (E2) + E2, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * E2 + E2, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * (E2) + E2, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * E2 + E2, sizeof(THING) , ...) | - kzalloc + kcalloc ( - (E1) * E2 + E1, E2 , ...) | - kzalloc + kcalloc ( - (E1) * (E2) + E1, E2 , ...) | - kzalloc + kcalloc ( - E1 * E2 + E1, E2 , ...) ) Signed-off-by: Kees Cook <keescook@chromium.org>
2018-06-13 05:03:40 +08:00
vpd_buf = kcalloc(entries, sizeof(unsigned long *), GFP_KERNEL);
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
if (!vpd_buf)
return -ENOMEM;
le = (struct sg_list *)get_zeroed_page(GFP_KERNEL);
if (!le) {
rc = -ENOMEM;
goto err1;
}
for (i = 0; i < entries; i++) {
vpd_buf[i] = (unsigned long *)get_zeroed_page(GFP_KERNEL);
if (!vpd_buf[i]) {
rc = -ENOMEM;
goto err2;
}
le[i].phys_addr = cpu_to_be64(virt_to_phys(vpd_buf[i]));
le[i].len = cpu_to_be64(SG_BUFFER_SIZE);
if ((i == (entries - 1)) && mod)
le[i].len = cpu_to_be64(mod);
}
if (adapter)
rc = cxl_h_collect_vpd_adapter(adapter->guest->handle,
virt_to_phys(le), entries, &out);
else
rc = cxl_h_collect_vpd(afu->guest->handle, 0,
virt_to_phys(le), entries, &out);
pr_devel("length of available (entries: %i), vpd: %#llx\n",
entries, out);
if (!rc) {
/*
* hcall returns in 'out' the size of available VPDs.
* It fills the buffer with as much data as possible.
*/
if (out < len)
len = out;
rc = len;
if (out) {
for (i = 0; i < entries; i++) {
if (len < SG_BUFFER_SIZE)
tocopy = len;
else
tocopy = SG_BUFFER_SIZE;
memcpy(buf, vpd_buf[i], tocopy);
buf += tocopy;
len -= tocopy;
}
}
}
err2:
for (i = 0; i < entries; i++) {
if (vpd_buf[i])
free_page((unsigned long) vpd_buf[i]);
}
free_page((unsigned long) le);
err1:
kfree(vpd_buf);
return rc;
}
static int guest_get_irq_info(struct cxl_context *ctx, struct cxl_irq_info *info)
{
return cxl_h_collect_int_info(ctx->afu->guest->handle, ctx->process_token, info);
}
static irqreturn_t guest_psl_irq(int irq, void *data)
{
struct cxl_context *ctx = data;
struct cxl_irq_info irq_info;
int rc;
pr_devel("%d: received PSL interrupt %i\n", ctx->pe, irq);
rc = guest_get_irq_info(ctx, &irq_info);
if (rc) {
WARN(1, "Unable to get IRQ info: %i\n", rc);
return IRQ_HANDLED;
}
cxl: Rename some psl8 specific functions 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> Reviewed-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com> Acked-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2017-04-07 22:11:57 +08:00
rc = cxl_irq_psl8(irq, ctx, &irq_info);
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
return rc;
}
cxl: Adapter failure handling Check the AFU state whenever an API is called. The hypervisor may issue a reset of the adapter when it detects a fault. When it happens, it launches an error recovery which will either move the AFU to a permanent failure state, or in the disabled state. If the AFU is found to be disabled, detach all existing contexts from it before issuing a AFU reset to re-enable it. Before detaching contexts, notify any kernel driver through the EEH callbacks of the AFU pci device. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:41 +08:00
static int afu_read_error_state(struct cxl_afu *afu, int *state_out)
{
u64 state;
int rc = 0;
cxl: Check periodically the coherent platform function's state In the PowerVM environment, the PHYP CoherentAccel component manages the state of the Coherent Accelerator Processor Interface adapter and virtualizes CAPI resources, handles CAPP, PSL, PSL Slice errors - and interrupts - and provides a new set of hcalls for the OS APIs to utilize Accelerator Function Unit (AFU). During the course of operation, a coherent platform function can encounter errors. Some possible reason for errors are: • Hardware recoverable and unrecoverable errors • Transient and over-threshold correctable errors PHYP implements its own state model for the coherent platform function. The state of the AFU is available through a hcall. The current implementation of the cxl driver, for the PowerVM environment, checks this state of the AFU only when an action is requested - open a device, ioctl command, memory map, attach/detach a process - from an external driver - cxlflash, libcxl. If an error is detected the cxl driver handles the error according the content of the Power Architecture Platform Requirements document. But in case of low-level troubles (or error injection), the PHYP component may reset the card and change the AFU state. The PHYP interface doesn't provide any way to be notified when that happens thus implies that the cxl driver: • cannot handle immediatly the state change of the AFU. • cannot notify other drivers (cxlflash, ...) The purpose of this patch is to wake up the cpu periodically to check the current state of each AFU and to see if we need to enter an error recovery path. Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-04-22 21:39:22 +08:00
if (!afu)
return -EIO;
cxl: Adapter failure handling Check the AFU state whenever an API is called. The hypervisor may issue a reset of the adapter when it detects a fault. When it happens, it launches an error recovery which will either move the AFU to a permanent failure state, or in the disabled state. If the AFU is found to be disabled, detach all existing contexts from it before issuing a AFU reset to re-enable it. Before detaching contexts, notify any kernel driver through the EEH callbacks of the AFU pci device. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:41 +08:00
rc = cxl_h_read_error_state(afu->guest->handle, &state);
if (!rc) {
WARN_ON(state != H_STATE_NORMAL &&
state != H_STATE_DISABLE &&
state != H_STATE_TEMP_UNAVAILABLE &&
state != H_STATE_PERM_UNAVAILABLE);
*state_out = state & 0xffffffff;
}
return rc;
}
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
static irqreturn_t guest_slice_irq_err(int irq, void *data)
{
struct cxl_afu *afu = data;
int rc;
cxl: Refine slice error debug messages The PSL Slice Error Register (PSL_SERR_An) reports implementation dependent AFU errors, in the form of a bitmap. The PSL_SERR_An register content is printed in the form of hex dump debug message. This patch decodes the PSL_ERR_An register contents, and prints a specific error message for each possible error bit. It also dumps the secondary registers AFU_ERR_An and PSL_DSISR_An, that may contain extra debug information. This patch also removes the large WARN message that used to report the cxl slice error interrupt, and replaces it by a short informative message, that draws attention to AFU implementation errors. Signed-off-by: Philippe Bergheaud <felix@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-07-05 19:08:06 +08:00
u64 serr, afu_error, dsisr;
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
rc = cxl_h_get_fn_error_interrupt(afu->guest->handle, &serr);
if (rc) {
dev_crit(&afu->dev, "Couldn't read PSL_SERR_An: %d\n", rc);
return IRQ_HANDLED;
}
cxl: Refine slice error debug messages The PSL Slice Error Register (PSL_SERR_An) reports implementation dependent AFU errors, in the form of a bitmap. The PSL_SERR_An register content is printed in the form of hex dump debug message. This patch decodes the PSL_ERR_An register contents, and prints a specific error message for each possible error bit. It also dumps the secondary registers AFU_ERR_An and PSL_DSISR_An, that may contain extra debug information. This patch also removes the large WARN message that used to report the cxl slice error interrupt, and replaces it by a short informative message, that draws attention to AFU implementation errors. Signed-off-by: Philippe Bergheaud <felix@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-07-05 19:08:06 +08:00
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, "AFU_ERR_An: 0x%.16llx\n", afu_error);
dev_crit(&afu->dev, "PSL_DSISR_An: 0x%.16llx\n", dsisr);
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
rc = cxl_h_ack_fn_error_interrupt(afu->guest->handle, serr);
if (rc)
dev_crit(&afu->dev, "Couldn't ack slice error interrupt: %d\n",
rc);
return IRQ_HANDLED;
}
static int irq_alloc_range(struct cxl *adapter, int len, int *irq)
{
int i, n;
struct irq_avail *cur;
for (i = 0; i < adapter->guest->irq_nranges; i++) {
cur = &adapter->guest->irq_avail[i];
n = bitmap_find_next_zero_area(cur->bitmap, cur->range,
0, len, 0);
if (n < cur->range) {
bitmap_set(cur->bitmap, n, len);
*irq = cur->offset + n;
pr_devel("guest: allocate IRQs %#x->%#x\n",
*irq, *irq + len - 1);
return 0;
}
}
return -ENOSPC;
}
static int irq_free_range(struct cxl *adapter, int irq, int len)
{
int i, n;
struct irq_avail *cur;
if (len == 0)
return -ENOENT;
for (i = 0; i < adapter->guest->irq_nranges; i++) {
cur = &adapter->guest->irq_avail[i];
if (irq >= cur->offset &&
(irq + len) <= (cur->offset + cur->range)) {
n = irq - cur->offset;
bitmap_clear(cur->bitmap, n, len);
pr_devel("guest: release IRQs %#x->%#x\n",
irq, irq + len - 1);
return 0;
}
}
return -ENOENT;
}
static int guest_reset(struct cxl *adapter)
{
cxl: Adapter failure handling Check the AFU state whenever an API is called. The hypervisor may issue a reset of the adapter when it detects a fault. When it happens, it launches an error recovery which will either move the AFU to a permanent failure state, or in the disabled state. If the AFU is found to be disabled, detach all existing contexts from it before issuing a AFU reset to re-enable it. Before detaching contexts, notify any kernel driver through the EEH callbacks of the AFU pci device. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:41 +08:00
struct cxl_afu *afu = NULL;
int i, rc;
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
pr_devel("Adapter reset request\n");
cxl: Wrap iterations over afu slices inside 'afu_list_lock' Within cxl module, iteration over array 'adapter->afu' may be racy at few points as it might be simultaneously read during an EEH and its contents being set to NULL while driver is being unloaded or unbound from the adapter. This might result in a NULL pointer to 'struct afu' being de-referenced during an EEH thereby causing a kernel oops. This patch fixes this by making sure that all access to the array 'adapter->afu' is wrapped within the context of spin-lock 'adapter->afu_list_lock'. Fixes: 9e8df8a21963 ("cxl: EEH support") Cc: stable@vger.kernel.org # v4.3+ Acked-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com> Acked-by: Frederic Barrat <fbarrat@linux.ibm.com> Acked-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Signed-off-by: Vaibhav Jain <vaibhav@linux.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2019-01-29 19:06:18 +08:00
spin_lock(&adapter->afu_list_lock);
cxl: Adapter failure handling Check the AFU state whenever an API is called. The hypervisor may issue a reset of the adapter when it detects a fault. When it happens, it launches an error recovery which will either move the AFU to a permanent failure state, or in the disabled state. If the AFU is found to be disabled, detach all existing contexts from it before issuing a AFU reset to re-enable it. Before detaching contexts, notify any kernel driver through the EEH callbacks of the AFU pci device. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:41 +08:00
for (i = 0; i < adapter->slices; i++) {
if ((afu = adapter->afu[i])) {
pci_error_handlers(afu, CXL_ERROR_DETECTED_EVENT,
pci_channel_io_frozen);
cxl_context_detach_all(afu);
}
}
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
rc = cxl_h_reset_adapter(adapter->guest->handle);
cxl: Adapter failure handling Check the AFU state whenever an API is called. The hypervisor may issue a reset of the adapter when it detects a fault. When it happens, it launches an error recovery which will either move the AFU to a permanent failure state, or in the disabled state. If the AFU is found to be disabled, detach all existing contexts from it before issuing a AFU reset to re-enable it. Before detaching contexts, notify any kernel driver through the EEH callbacks of the AFU pci device. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:41 +08:00
for (i = 0; i < adapter->slices; i++) {
if (!rc && (afu = adapter->afu[i])) {
pci_error_handlers(afu, CXL_SLOT_RESET_EVENT,
pci_channel_io_normal);
pci_error_handlers(afu, CXL_RESUME_EVENT, 0);
}
}
cxl: Wrap iterations over afu slices inside 'afu_list_lock' Within cxl module, iteration over array 'adapter->afu' may be racy at few points as it might be simultaneously read during an EEH and its contents being set to NULL while driver is being unloaded or unbound from the adapter. This might result in a NULL pointer to 'struct afu' being de-referenced during an EEH thereby causing a kernel oops. This patch fixes this by making sure that all access to the array 'adapter->afu' is wrapped within the context of spin-lock 'adapter->afu_list_lock'. Fixes: 9e8df8a21963 ("cxl: EEH support") Cc: stable@vger.kernel.org # v4.3+ Acked-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com> Acked-by: Frederic Barrat <fbarrat@linux.ibm.com> Acked-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Signed-off-by: Vaibhav Jain <vaibhav@linux.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2019-01-29 19:06:18 +08:00
spin_unlock(&adapter->afu_list_lock);
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
return rc;
}
static int guest_alloc_one_irq(struct cxl *adapter)
{
int irq;
spin_lock(&adapter->guest->irq_alloc_lock);
if (irq_alloc_range(adapter, 1, &irq))
irq = -ENOSPC;
spin_unlock(&adapter->guest->irq_alloc_lock);
return irq;
}
static void guest_release_one_irq(struct cxl *adapter, int irq)
{
spin_lock(&adapter->guest->irq_alloc_lock);
irq_free_range(adapter, irq, 1);
spin_unlock(&adapter->guest->irq_alloc_lock);
}
static int guest_alloc_irq_ranges(struct cxl_irq_ranges *irqs,
struct cxl *adapter, unsigned int num)
{
int i, try, irq;
memset(irqs, 0, sizeof(struct cxl_irq_ranges));
spin_lock(&adapter->guest->irq_alloc_lock);
for (i = 0; i < CXL_IRQ_RANGES && num; i++) {
try = num;
while (try) {
if (irq_alloc_range(adapter, try, &irq) == 0)
break;
try /= 2;
}
if (!try)
goto error;
irqs->offset[i] = irq;
irqs->range[i] = try;
num -= try;
}
if (num)
goto error;
spin_unlock(&adapter->guest->irq_alloc_lock);
return 0;
error:
for (i = 0; i < CXL_IRQ_RANGES; i++)
irq_free_range(adapter, irqs->offset[i], irqs->range[i]);
spin_unlock(&adapter->guest->irq_alloc_lock);
return -ENOSPC;
}
static void guest_release_irq_ranges(struct cxl_irq_ranges *irqs,
struct cxl *adapter)
{
int i;
spin_lock(&adapter->guest->irq_alloc_lock);
for (i = 0; i < CXL_IRQ_RANGES; i++)
irq_free_range(adapter, irqs->offset[i], irqs->range[i]);
spin_unlock(&adapter->guest->irq_alloc_lock);
}
static int guest_register_serr_irq(struct cxl_afu *afu)
{
afu->err_irq_name = kasprintf(GFP_KERNEL, "cxl-%s-err",
dev_name(&afu->dev));
if (!afu->err_irq_name)
return -ENOMEM;
if (!(afu->serr_virq = cxl_map_irq(afu->adapter, afu->serr_hwirq,
guest_slice_irq_err, afu, afu->err_irq_name))) {
kfree(afu->err_irq_name);
afu->err_irq_name = NULL;
return -ENOMEM;
}
return 0;
}
static void guest_release_serr_irq(struct cxl_afu *afu)
{
cxl_unmap_irq(afu->serr_virq, afu);
cxl_ops->release_one_irq(afu->adapter, afu->serr_hwirq);
kfree(afu->err_irq_name);
}
static int guest_ack_irq(struct cxl_context *ctx, u64 tfc, u64 psl_reset_mask)
{
return cxl_h_control_faults(ctx->afu->guest->handle, ctx->process_token,
tfc >> 32, (psl_reset_mask != 0));
}
static void disable_afu_irqs(struct cxl_context *ctx)
{
irq_hw_number_t hwirq;
unsigned int virq;
int r, i;
pr_devel("Disabling AFU(%d) interrupts\n", ctx->afu->slice);
for (r = 0; r < CXL_IRQ_RANGES; r++) {
hwirq = ctx->irqs.offset[r];
for (i = 0; i < ctx->irqs.range[r]; hwirq++, i++) {
virq = irq_find_mapping(NULL, hwirq);
disable_irq(virq);
}
}
}
static void enable_afu_irqs(struct cxl_context *ctx)
{
irq_hw_number_t hwirq;
unsigned int virq;
int r, i;
pr_devel("Enabling AFU(%d) interrupts\n", ctx->afu->slice);
for (r = 0; r < CXL_IRQ_RANGES; r++) {
hwirq = ctx->irqs.offset[r];
for (i = 0; i < ctx->irqs.range[r]; hwirq++, i++) {
virq = irq_find_mapping(NULL, hwirq);
enable_irq(virq);
}
}
}
static int _guest_afu_cr_readXX(int sz, struct cxl_afu *afu, int cr_idx,
u64 offset, u64 *val)
{
unsigned long cr;
char c;
int rc = 0;
if (afu->crs_len < sz)
return -ENOENT;
if (unlikely(offset >= afu->crs_len))
return -ERANGE;
cr = get_zeroed_page(GFP_KERNEL);
if (!cr)
return -ENOMEM;
rc = cxl_h_get_config(afu->guest->handle, cr_idx, offset,
virt_to_phys((void *)cr), sz);
if (rc)
goto err;
switch (sz) {
case 1:
c = *((char *) cr);
*val = c;
break;
case 2:
*val = in_le16((u16 *)cr);
break;
case 4:
*val = in_le32((unsigned *)cr);
break;
case 8:
*val = in_le64((u64 *)cr);
break;
default:
WARN_ON(1);
}
err:
free_page(cr);
return rc;
}
static int guest_afu_cr_read32(struct cxl_afu *afu, int cr_idx, u64 offset,
u32 *out)
{
int rc;
u64 val;
rc = _guest_afu_cr_readXX(4, afu, cr_idx, offset, &val);
if (!rc)
*out = (u32) val;
return rc;
}
static int guest_afu_cr_read16(struct cxl_afu *afu, int cr_idx, u64 offset,
u16 *out)
{
int rc;
u64 val;
rc = _guest_afu_cr_readXX(2, afu, cr_idx, offset, &val);
if (!rc)
*out = (u16) val;
return rc;
}
static int guest_afu_cr_read8(struct cxl_afu *afu, int cr_idx, u64 offset,
u8 *out)
{
int rc;
u64 val;
rc = _guest_afu_cr_readXX(1, afu, cr_idx, offset, &val);
if (!rc)
*out = (u8) val;
return rc;
}
static int guest_afu_cr_read64(struct cxl_afu *afu, int cr_idx, u64 offset,
u64 *out)
{
return _guest_afu_cr_readXX(8, afu, cr_idx, offset, out);
}
cxl: Support the cxl kernel API from a guest Like on bare-metal, the cxl driver creates a virtual PHB and a pci device for the AFU. The configuration space of the device is mapped to the configuration record of the AFU. Reuse the code defined in afu_cr_read8|16|32() when reading the configuration space of the AFU device. Even though the (virtual) AFU device is a pci device, the adapter is not. So a driver using the cxl kernel API cannot read the VPD of the adapter through the usual PCI interface. Therefore, we add a call to the cxl kernel API: ssize_t cxl_read_adapter_vpd(struct pci_dev *dev, void *buf, size_t count); Co-authored-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:40 +08:00
static int guest_afu_cr_write32(struct cxl_afu *afu, int cr, u64 off, u32 in)
{
/* config record is not writable from guest */
return -EPERM;
}
static int guest_afu_cr_write16(struct cxl_afu *afu, int cr, u64 off, u16 in)
{
/* config record is not writable from guest */
return -EPERM;
}
static int guest_afu_cr_write8(struct cxl_afu *afu, int cr, u64 off, u8 in)
{
/* config record is not writable from guest */
return -EPERM;
}
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
static int attach_afu_directed(struct cxl_context *ctx, u64 wed, u64 amr)
{
struct cxl_process_element_hcall *elem;
struct cxl *adapter = ctx->afu->adapter;
const struct cred *cred;
u32 pid, idx;
int rc, r, i;
u64 mmio_addr, mmio_size;
__be64 flags = 0;
/* Must be 8 byte aligned and cannot cross a 4096 byte boundary */
if (!(elem = (struct cxl_process_element_hcall *)
get_zeroed_page(GFP_KERNEL)))
return -ENOMEM;
elem->version = cpu_to_be64(CXL_PROCESS_ELEMENT_VERSION);
if (ctx->kernel) {
pid = 0;
flags |= CXL_PE_TRANSLATION_ENABLED;
flags |= CXL_PE_PRIVILEGED_PROCESS;
if (mfmsr() & MSR_SF)
flags |= CXL_PE_64_BIT;
} else {
pid = current->pid;
flags |= CXL_PE_PROBLEM_STATE;
flags |= CXL_PE_TRANSLATION_ENABLED;
if (!test_tsk_thread_flag(current, TIF_32BIT))
flags |= CXL_PE_64_BIT;
cred = get_current_cred();
if (uid_eq(cred->euid, GLOBAL_ROOT_UID))
flags |= CXL_PE_PRIVILEGED_PROCESS;
put_cred(cred);
}
elem->flags = cpu_to_be64(flags);
elem->common.tid = cpu_to_be32(0); /* Unused */
elem->common.pid = cpu_to_be32(pid);
elem->common.csrp = cpu_to_be64(0); /* disable */
cxl: Add psl9 specific code 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> Acked-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> [mpe: Fixup comment formatting, unsplit long strings] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2017-04-12 22:34:07 +08:00
elem->common.u.psl8.aurp0 = cpu_to_be64(0); /* disable */
elem->common.u.psl8.aurp1 = cpu_to_be64(0); /* disable */
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
cxl_prefault(ctx, wed);
cxl: Add psl9 specific code 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> Acked-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> [mpe: Fixup comment formatting, unsplit long strings] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2017-04-12 22:34:07 +08:00
elem->common.u.psl8.sstp0 = cpu_to_be64(ctx->sstp0);
elem->common.u.psl8.sstp1 = cpu_to_be64(ctx->sstp1);
cxl: Ensure PSL interrupt is configured for contexts with no AFU IRQs In the cxl kernel API, it is possible to create a context and start it without allocating any interrupts. Since we assign or allocate the PSL interrupt when allocating AFU interrupts this will lead to a situation where we start the context with no means to take any faults. The user API is not affected as it always goes through the cxl interrupt allocation code paths and will have the PSL interrupt allocated or assigned, even if no AFU interrupts were requested. This checks that at least one interrupt is configured at the time of attach, and if not it will assign the multiplexed PSL interrupt for powernv, or allocate a single interrupt for PowerVM. Signed-off-by: Ian Munsie <imunsie@au1.ibm.com> Reviewed-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-05-04 12:52:58 +08:00
/*
* Ensure we have at least one interrupt allocated to take faults for
* kernel contexts that may not have allocated any AFU IRQs at all:
*/
if (ctx->irqs.range[0] == 0) {
rc = afu_register_irqs(ctx, 0);
if (rc)
goto out_free;
}
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
for (r = 0; r < CXL_IRQ_RANGES; r++) {
for (i = 0; i < ctx->irqs.range[r]; i++) {
if (r == 0 && i == 0) {
elem->pslVirtualIsn = cpu_to_be32(ctx->irqs.offset[0]);
} else {
idx = ctx->irqs.offset[r] + i - adapter->guest->irq_base_offset;
elem->applicationVirtualIsnBitmap[idx / 8] |= 0x80 >> (idx % 8);
}
}
}
elem->common.amr = cpu_to_be64(amr);
elem->common.wed = cpu_to_be64(wed);
disable_afu_irqs(ctx);
rc = cxl_h_attach_process(ctx->afu->guest->handle, elem,
&ctx->process_token, &mmio_addr, &mmio_size);
if (rc == H_SUCCESS) {
if (ctx->master || !ctx->afu->pp_psa) {
ctx->psn_phys = ctx->afu->psn_phys;
ctx->psn_size = ctx->afu->adapter->ps_size;
} else {
ctx->psn_phys = mmio_addr;
ctx->psn_size = mmio_size;
}
if (ctx->afu->pp_psa && mmio_size &&
ctx->afu->pp_size == 0) {
/*
* There's no property in the device tree to read the
* pp_size. We only find out at the 1st attach.
* Compared to bare-metal, it is too late and we
* should really lock here. However, on powerVM,
* pp_size is really only used to display in /sys.
* Being discussed with pHyp for their next release.
*/
ctx->afu->pp_size = mmio_size;
}
/* from PAPR: process element is bytes 4-7 of process token */
ctx->external_pe = ctx->process_token & 0xFFFFFFFF;
pr_devel("CXL pe=%i is known as %i for pHyp, mmio_size=%#llx",
ctx->pe, ctx->external_pe, ctx->psn_size);
ctx->pe_inserted = true;
enable_afu_irqs(ctx);
}
cxl: Ensure PSL interrupt is configured for contexts with no AFU IRQs In the cxl kernel API, it is possible to create a context and start it without allocating any interrupts. Since we assign or allocate the PSL interrupt when allocating AFU interrupts this will lead to a situation where we start the context with no means to take any faults. The user API is not affected as it always goes through the cxl interrupt allocation code paths and will have the PSL interrupt allocated or assigned, even if no AFU interrupts were requested. This checks that at least one interrupt is configured at the time of attach, and if not it will assign the multiplexed PSL interrupt for powernv, or allocate a single interrupt for PowerVM. Signed-off-by: Ian Munsie <imunsie@au1.ibm.com> Reviewed-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-05-04 12:52:58 +08:00
out_free:
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
free_page((u64)elem);
return rc;
}
static int guest_attach_process(struct cxl_context *ctx, bool kernel, u64 wed, u64 amr)
{
pr_devel("in %s\n", __func__);
ctx->kernel = kernel;
if (ctx->afu->current_mode == CXL_MODE_DIRECTED)
return attach_afu_directed(ctx, wed, amr);
/* dedicated mode not supported on FW840 */
return -EINVAL;
}
static int detach_afu_directed(struct cxl_context *ctx)
{
if (!ctx->pe_inserted)
return 0;
if (cxl_h_detach_process(ctx->afu->guest->handle, ctx->process_token))
return -1;
return 0;
}
static int guest_detach_process(struct cxl_context *ctx)
{
pr_devel("in %s\n", __func__);
trace_cxl_detach(ctx);
cxl: Adapter failure handling Check the AFU state whenever an API is called. The hypervisor may issue a reset of the adapter when it detects a fault. When it happens, it launches an error recovery which will either move the AFU to a permanent failure state, or in the disabled state. If the AFU is found to be disabled, detach all existing contexts from it before issuing a AFU reset to re-enable it. Before detaching contexts, notify any kernel driver through the EEH callbacks of the AFU pci device. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:41 +08:00
if (!cxl_ops->link_ok(ctx->afu->adapter, ctx->afu))
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
return -EIO;
if (ctx->afu->current_mode == CXL_MODE_DIRECTED)
return detach_afu_directed(ctx);
return -EINVAL;
}
static void guest_release_afu(struct device *dev)
{
struct cxl_afu *afu = to_cxl_afu(dev);
pr_devel("%s\n", __func__);
idr_destroy(&afu->contexts_idr);
kfree(afu->guest);
kfree(afu);
}
ssize_t cxl_guest_read_afu_vpd(struct cxl_afu *afu, void *buf, size_t len)
{
return guest_collect_vpd(NULL, afu, buf, len);
}
#define ERR_BUFF_MAX_COPY_SIZE PAGE_SIZE
static ssize_t guest_afu_read_err_buffer(struct cxl_afu *afu, char *buf,
loff_t off, size_t count)
{
void *tbuf = NULL;
int rc = 0;
tbuf = (void *) get_zeroed_page(GFP_KERNEL);
if (!tbuf)
return -ENOMEM;
rc = cxl_h_get_afu_err(afu->guest->handle,
off & 0x7,
virt_to_phys(tbuf),
count);
if (rc)
goto err;
if (count > ERR_BUFF_MAX_COPY_SIZE)
count = ERR_BUFF_MAX_COPY_SIZE - (off & 0x7);
memcpy(buf, tbuf, count);
err:
free_page((u64)tbuf);
return rc;
}
static int guest_afu_check_and_enable(struct cxl_afu *afu)
{
return 0;
}
cxl: sysfs support for guests Filter out a few adapter parameters which don't make sense in a guest. Document the changes. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:37 +08:00
static bool guest_support_attributes(const char *attr_name,
enum cxl_attrs type)
{
switch (type) {
case CXL_ADAPTER_ATTRS:
if ((strcmp(attr_name, "base_image") == 0) ||
(strcmp(attr_name, "load_image_on_perst") == 0) ||
(strcmp(attr_name, "perst_reloads_same_image") == 0) ||
(strcmp(attr_name, "image_loaded") == 0))
return false;
break;
case CXL_AFU_MASTER_ATTRS:
if ((strcmp(attr_name, "pp_mmio_off") == 0))
return false;
break;
case CXL_AFU_ATTRS:
break;
default:
break;
}
return true;
}
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
static int activate_afu_directed(struct cxl_afu *afu)
{
int rc;
dev_info(&afu->dev, "Activating AFU(%d) directed mode\n", afu->slice);
afu->current_mode = CXL_MODE_DIRECTED;
afu->num_procs = afu->max_procs_virtualised;
if ((rc = cxl_chardev_m_afu_add(afu)))
return rc;
if ((rc = cxl_sysfs_afu_m_add(afu)))
goto err;
if ((rc = cxl_chardev_s_afu_add(afu)))
goto err1;
return 0;
err1:
cxl_sysfs_afu_m_remove(afu);
err:
cxl_chardev_afu_remove(afu);
return rc;
}
static int guest_afu_activate_mode(struct cxl_afu *afu, int mode)
{
if (!mode)
return 0;
if (!(mode & afu->modes_supported))
return -EINVAL;
if (mode == CXL_MODE_DIRECTED)
return activate_afu_directed(afu);
if (mode == CXL_MODE_DEDICATED)
dev_err(&afu->dev, "Dedicated mode not supported\n");
return -EINVAL;
}
static int deactivate_afu_directed(struct cxl_afu *afu)
{
dev_info(&afu->dev, "Deactivating AFU(%d) directed mode\n", afu->slice);
afu->current_mode = 0;
afu->num_procs = 0;
cxl_sysfs_afu_m_remove(afu);
cxl_chardev_afu_remove(afu);
cxl_ops->afu_reset(afu);
return 0;
}
static int guest_afu_deactivate_mode(struct cxl_afu *afu, int mode)
{
if (!mode)
return 0;
if (!(mode & afu->modes_supported))
return -EINVAL;
if (mode == CXL_MODE_DIRECTED)
return deactivate_afu_directed(afu);
return 0;
}
static int guest_afu_reset(struct cxl_afu *afu)
{
pr_devel("AFU(%d) reset request\n", afu->slice);
return cxl_h_reset_afu(afu->guest->handle);
}
static int guest_map_slice_regs(struct cxl_afu *afu)
{
if (!(afu->p2n_mmio = ioremap(afu->guest->p2n_phys, afu->guest->p2n_size))) {
dev_err(&afu->dev, "Error mapping AFU(%d) MMIO regions\n",
afu->slice);
return -ENOMEM;
}
return 0;
}
static void guest_unmap_slice_regs(struct cxl_afu *afu)
{
if (afu->p2n_mmio)
iounmap(afu->p2n_mmio);
}
cxl: Adapter failure handling Check the AFU state whenever an API is called. The hypervisor may issue a reset of the adapter when it detects a fault. When it happens, it launches an error recovery which will either move the AFU to a permanent failure state, or in the disabled state. If the AFU is found to be disabled, detach all existing contexts from it before issuing a AFU reset to re-enable it. Before detaching contexts, notify any kernel driver through the EEH callbacks of the AFU pci device. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:41 +08:00
static int afu_update_state(struct cxl_afu *afu)
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
{
cxl: Adapter failure handling Check the AFU state whenever an API is called. The hypervisor may issue a reset of the adapter when it detects a fault. When it happens, it launches an error recovery which will either move the AFU to a permanent failure state, or in the disabled state. If the AFU is found to be disabled, detach all existing contexts from it before issuing a AFU reset to re-enable it. Before detaching contexts, notify any kernel driver through the EEH callbacks of the AFU pci device. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:41 +08:00
int rc, cur_state;
rc = afu_read_error_state(afu, &cur_state);
if (rc)
return rc;
if (afu->guest->previous_state == cur_state)
return 0;
pr_devel("AFU(%d) update state to %#x\n", afu->slice, cur_state);
switch (cur_state) {
case H_STATE_NORMAL:
afu->guest->previous_state = cur_state;
break;
case H_STATE_DISABLE:
pci_error_handlers(afu, CXL_ERROR_DETECTED_EVENT,
pci_channel_io_frozen);
cxl_context_detach_all(afu);
if ((rc = cxl_ops->afu_reset(afu)))
pr_devel("reset hcall failed %d\n", rc);
rc = afu_read_error_state(afu, &cur_state);
if (!rc && cur_state == H_STATE_NORMAL) {
pci_error_handlers(afu, CXL_SLOT_RESET_EVENT,
pci_channel_io_normal);
pci_error_handlers(afu, CXL_RESUME_EVENT, 0);
}
afu->guest->previous_state = 0;
break;
case H_STATE_TEMP_UNAVAILABLE:
afu->guest->previous_state = cur_state;
break;
case H_STATE_PERM_UNAVAILABLE:
dev_err(&afu->dev, "AFU is in permanent error state\n");
pci_error_handlers(afu, CXL_ERROR_DETECTED_EVENT,
pci_channel_io_perm_failure);
afu->guest->previous_state = cur_state;
break;
default:
pr_err("Unexpected AFU(%d) error state: %#x\n",
afu->slice, cur_state);
return -EINVAL;
}
return rc;
}
cxl: Check periodically the coherent platform function's state In the PowerVM environment, the PHYP CoherentAccel component manages the state of the Coherent Accelerator Processor Interface adapter and virtualizes CAPI resources, handles CAPP, PSL, PSL Slice errors - and interrupts - and provides a new set of hcalls for the OS APIs to utilize Accelerator Function Unit (AFU). During the course of operation, a coherent platform function can encounter errors. Some possible reason for errors are: • Hardware recoverable and unrecoverable errors • Transient and over-threshold correctable errors PHYP implements its own state model for the coherent platform function. The state of the AFU is available through a hcall. The current implementation of the cxl driver, for the PowerVM environment, checks this state of the AFU only when an action is requested - open a device, ioctl command, memory map, attach/detach a process - from an external driver - cxlflash, libcxl. If an error is detected the cxl driver handles the error according the content of the Power Architecture Platform Requirements document. But in case of low-level troubles (or error injection), the PHYP component may reset the card and change the AFU state. The PHYP interface doesn't provide any way to be notified when that happens thus implies that the cxl driver: • cannot handle immediatly the state change of the AFU. • cannot notify other drivers (cxlflash, ...) The purpose of this patch is to wake up the cpu periodically to check the current state of each AFU and to see if we need to enter an error recovery path. Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-04-22 21:39:22 +08:00
static void afu_handle_errstate(struct work_struct *work)
cxl: Adapter failure handling Check the AFU state whenever an API is called. The hypervisor may issue a reset of the adapter when it detects a fault. When it happens, it launches an error recovery which will either move the AFU to a permanent failure state, or in the disabled state. If the AFU is found to be disabled, detach all existing contexts from it before issuing a AFU reset to re-enable it. Before detaching contexts, notify any kernel driver through the EEH callbacks of the AFU pci device. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:41 +08:00
{
cxl: Check periodically the coherent platform function's state In the PowerVM environment, the PHYP CoherentAccel component manages the state of the Coherent Accelerator Processor Interface adapter and virtualizes CAPI resources, handles CAPP, PSL, PSL Slice errors - and interrupts - and provides a new set of hcalls for the OS APIs to utilize Accelerator Function Unit (AFU). During the course of operation, a coherent platform function can encounter errors. Some possible reason for errors are: • Hardware recoverable and unrecoverable errors • Transient and over-threshold correctable errors PHYP implements its own state model for the coherent platform function. The state of the AFU is available through a hcall. The current implementation of the cxl driver, for the PowerVM environment, checks this state of the AFU only when an action is requested - open a device, ioctl command, memory map, attach/detach a process - from an external driver - cxlflash, libcxl. If an error is detected the cxl driver handles the error according the content of the Power Architecture Platform Requirements document. But in case of low-level troubles (or error injection), the PHYP component may reset the card and change the AFU state. The PHYP interface doesn't provide any way to be notified when that happens thus implies that the cxl driver: • cannot handle immediatly the state change of the AFU. • cannot notify other drivers (cxlflash, ...) The purpose of this patch is to wake up the cpu periodically to check the current state of each AFU and to see if we need to enter an error recovery path. Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-04-22 21:39:22 +08:00
struct cxl_afu_guest *afu_guest =
container_of(to_delayed_work(work), struct cxl_afu_guest, work_err);
cxl: Adapter failure handling Check the AFU state whenever an API is called. The hypervisor may issue a reset of the adapter when it detects a fault. When it happens, it launches an error recovery which will either move the AFU to a permanent failure state, or in the disabled state. If the AFU is found to be disabled, detach all existing contexts from it before issuing a AFU reset to re-enable it. Before detaching contexts, notify any kernel driver through the EEH callbacks of the AFU pci device. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:41 +08:00
cxl: Check periodically the coherent platform function's state In the PowerVM environment, the PHYP CoherentAccel component manages the state of the Coherent Accelerator Processor Interface adapter and virtualizes CAPI resources, handles CAPP, PSL, PSL Slice errors - and interrupts - and provides a new set of hcalls for the OS APIs to utilize Accelerator Function Unit (AFU). During the course of operation, a coherent platform function can encounter errors. Some possible reason for errors are: • Hardware recoverable and unrecoverable errors • Transient and over-threshold correctable errors PHYP implements its own state model for the coherent platform function. The state of the AFU is available through a hcall. The current implementation of the cxl driver, for the PowerVM environment, checks this state of the AFU only when an action is requested - open a device, ioctl command, memory map, attach/detach a process - from an external driver - cxlflash, libcxl. If an error is detected the cxl driver handles the error according the content of the Power Architecture Platform Requirements document. But in case of low-level troubles (or error injection), the PHYP component may reset the card and change the AFU state. The PHYP interface doesn't provide any way to be notified when that happens thus implies that the cxl driver: • cannot handle immediatly the state change of the AFU. • cannot notify other drivers (cxlflash, ...) The purpose of this patch is to wake up the cpu periodically to check the current state of each AFU and to see if we need to enter an error recovery path. Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-04-22 21:39:22 +08:00
if (!afu_update_state(afu_guest->parent) &&
afu_guest->previous_state == H_STATE_PERM_UNAVAILABLE)
return;
cxl: Fix coccinelle warnings Fix the following coccinelle warnings: drivers/misc/cxl/debugfs.c:46:0-23: WARNING: fops_io_x64 should be defined with DEFINE_DEBUGFS_ATTRIBUTE drivers/misc/cxl/guest.c:890:5-26: WARNING: Comparison to bool drivers/misc/cxl/irq.c:107:3-23: WARNING: Assignment of bool to 0/1 drivers/misc/cxl/native.c:57:2-3: Unneeded semicolon drivers/misc/cxl/native.c:170:2-3: Unneeded semicolon Signed-off-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com> Acked-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Reviewed-by: Matthew R. Ochs <mrochs@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-11-22 18:13:27 +08:00
if (afu_guest->handle_err)
cxl: Check periodically the coherent platform function's state In the PowerVM environment, the PHYP CoherentAccel component manages the state of the Coherent Accelerator Processor Interface adapter and virtualizes CAPI resources, handles CAPP, PSL, PSL Slice errors - and interrupts - and provides a new set of hcalls for the OS APIs to utilize Accelerator Function Unit (AFU). During the course of operation, a coherent platform function can encounter errors. Some possible reason for errors are: • Hardware recoverable and unrecoverable errors • Transient and over-threshold correctable errors PHYP implements its own state model for the coherent platform function. The state of the AFU is available through a hcall. The current implementation of the cxl driver, for the PowerVM environment, checks this state of the AFU only when an action is requested - open a device, ioctl command, memory map, attach/detach a process - from an external driver - cxlflash, libcxl. If an error is detected the cxl driver handles the error according the content of the Power Architecture Platform Requirements document. But in case of low-level troubles (or error injection), the PHYP component may reset the card and change the AFU state. The PHYP interface doesn't provide any way to be notified when that happens thus implies that the cxl driver: • cannot handle immediatly the state change of the AFU. • cannot notify other drivers (cxlflash, ...) The purpose of this patch is to wake up the cpu periodically to check the current state of each AFU and to see if we need to enter an error recovery path. Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-04-22 21:39:22 +08:00
schedule_delayed_work(&afu_guest->work_err,
msecs_to_jiffies(3000));
cxl: Adapter failure handling Check the AFU state whenever an API is called. The hypervisor may issue a reset of the adapter when it detects a fault. When it happens, it launches an error recovery which will either move the AFU to a permanent failure state, or in the disabled state. If the AFU is found to be disabled, detach all existing contexts from it before issuing a AFU reset to re-enable it. Before detaching contexts, notify any kernel driver through the EEH callbacks of the AFU pci device. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:41 +08:00
}
static bool guest_link_ok(struct cxl *cxl, struct cxl_afu *afu)
{
int state;
cxl: Check periodically the coherent platform function's state In the PowerVM environment, the PHYP CoherentAccel component manages the state of the Coherent Accelerator Processor Interface adapter and virtualizes CAPI resources, handles CAPP, PSL, PSL Slice errors - and interrupts - and provides a new set of hcalls for the OS APIs to utilize Accelerator Function Unit (AFU). During the course of operation, a coherent platform function can encounter errors. Some possible reason for errors are: • Hardware recoverable and unrecoverable errors • Transient and over-threshold correctable errors PHYP implements its own state model for the coherent platform function. The state of the AFU is available through a hcall. The current implementation of the cxl driver, for the PowerVM environment, checks this state of the AFU only when an action is requested - open a device, ioctl command, memory map, attach/detach a process - from an external driver - cxlflash, libcxl. If an error is detected the cxl driver handles the error according the content of the Power Architecture Platform Requirements document. But in case of low-level troubles (or error injection), the PHYP component may reset the card and change the AFU state. The PHYP interface doesn't provide any way to be notified when that happens thus implies that the cxl driver: • cannot handle immediatly the state change of the AFU. • cannot notify other drivers (cxlflash, ...) The purpose of this patch is to wake up the cpu periodically to check the current state of each AFU and to see if we need to enter an error recovery path. Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-04-22 21:39:22 +08:00
if (afu && (!afu_read_error_state(afu, &state))) {
if (state == H_STATE_NORMAL)
return true;
cxl: Adapter failure handling Check the AFU state whenever an API is called. The hypervisor may issue a reset of the adapter when it detects a fault. When it happens, it launches an error recovery which will either move the AFU to a permanent failure state, or in the disabled state. If the AFU is found to be disabled, detach all existing contexts from it before issuing a AFU reset to re-enable it. Before detaching contexts, notify any kernel driver through the EEH callbacks of the AFU pci device. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:41 +08:00
}
cxl: Check periodically the coherent platform function's state In the PowerVM environment, the PHYP CoherentAccel component manages the state of the Coherent Accelerator Processor Interface adapter and virtualizes CAPI resources, handles CAPP, PSL, PSL Slice errors - and interrupts - and provides a new set of hcalls for the OS APIs to utilize Accelerator Function Unit (AFU). During the course of operation, a coherent platform function can encounter errors. Some possible reason for errors are: • Hardware recoverable and unrecoverable errors • Transient and over-threshold correctable errors PHYP implements its own state model for the coherent platform function. The state of the AFU is available through a hcall. The current implementation of the cxl driver, for the PowerVM environment, checks this state of the AFU only when an action is requested - open a device, ioctl command, memory map, attach/detach a process - from an external driver - cxlflash, libcxl. If an error is detected the cxl driver handles the error according the content of the Power Architecture Platform Requirements document. But in case of low-level troubles (or error injection), the PHYP component may reset the card and change the AFU state. The PHYP interface doesn't provide any way to be notified when that happens thus implies that the cxl driver: • cannot handle immediatly the state change of the AFU. • cannot notify other drivers (cxlflash, ...) The purpose of this patch is to wake up the cpu periodically to check the current state of each AFU and to see if we need to enter an error recovery path. Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-04-22 21:39:22 +08:00
return false;
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
}
static int afu_properties_look_ok(struct cxl_afu *afu)
{
if (afu->pp_irqs < 0) {
dev_err(&afu->dev, "Unexpected per-process minimum interrupt value\n");
return -EINVAL;
}
if (afu->max_procs_virtualised < 1) {
dev_err(&afu->dev, "Unexpected max number of processes virtualised value\n");
return -EINVAL;
}
return 0;
}
int cxl_guest_init_afu(struct cxl *adapter, int slice, struct device_node *afu_np)
{
struct cxl_afu *afu;
bool free = true;
int rc;
pr_devel("in %s - AFU(%d)\n", __func__, slice);
if (!(afu = cxl_alloc_afu(adapter, slice)))
return -ENOMEM;
if (!(afu->guest = kzalloc(sizeof(struct cxl_afu_guest), GFP_KERNEL))) {
kfree(afu);
return -ENOMEM;
}
if ((rc = dev_set_name(&afu->dev, "afu%i.%i",
adapter->adapter_num,
slice)))
goto err1;
adapter->slices++;
if ((rc = cxl_of_read_afu_handle(afu, afu_np)))
goto err1;
if ((rc = cxl_ops->afu_reset(afu)))
goto err1;
if ((rc = cxl_of_read_afu_properties(afu, afu_np)))
goto err1;
if ((rc = afu_properties_look_ok(afu)))
goto err1;
if ((rc = guest_map_slice_regs(afu)))
goto err1;
if ((rc = guest_register_serr_irq(afu)))
goto err2;
/*
* After we call this function we must not free the afu directly, even
* if it returns an error!
*/
if ((rc = cxl_register_afu(afu)))
goto err_put1;
if ((rc = cxl_sysfs_afu_add(afu)))
goto err_put1;
/*
* pHyp doesn't expose the programming models supported by the
* AFU. pHyp currently only supports directed mode. If it adds
* dedicated mode later, this version of cxl has no way to
* detect it. So we'll initialize the driver, but the first
* attach will fail.
* Being discussed with pHyp to do better (likely new property)
*/
if (afu->max_procs_virtualised == 1)
afu->modes_supported = CXL_MODE_DEDICATED;
else
afu->modes_supported = CXL_MODE_DIRECTED;
if ((rc = cxl_afu_select_best_mode(afu)))
goto err_put2;
adapter->afu[afu->slice] = afu;
afu->enabled = true;
cxl: Check periodically the coherent platform function's state In the PowerVM environment, the PHYP CoherentAccel component manages the state of the Coherent Accelerator Processor Interface adapter and virtualizes CAPI resources, handles CAPP, PSL, PSL Slice errors - and interrupts - and provides a new set of hcalls for the OS APIs to utilize Accelerator Function Unit (AFU). During the course of operation, a coherent platform function can encounter errors. Some possible reason for errors are: • Hardware recoverable and unrecoverable errors • Transient and over-threshold correctable errors PHYP implements its own state model for the coherent platform function. The state of the AFU is available through a hcall. The current implementation of the cxl driver, for the PowerVM environment, checks this state of the AFU only when an action is requested - open a device, ioctl command, memory map, attach/detach a process - from an external driver - cxlflash, libcxl. If an error is detected the cxl driver handles the error according the content of the Power Architecture Platform Requirements document. But in case of low-level troubles (or error injection), the PHYP component may reset the card and change the AFU state. The PHYP interface doesn't provide any way to be notified when that happens thus implies that the cxl driver: • cannot handle immediatly the state change of the AFU. • cannot notify other drivers (cxlflash, ...) The purpose of this patch is to wake up the cpu periodically to check the current state of each AFU and to see if we need to enter an error recovery path. Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-04-22 21:39:22 +08:00
/*
* wake up the cpu periodically to check the state
* of the AFU using "afu" stored in the guest structure.
*/
afu->guest->parent = afu;
afu->guest->handle_err = true;
INIT_DELAYED_WORK(&afu->guest->work_err, afu_handle_errstate);
schedule_delayed_work(&afu->guest->work_err, msecs_to_jiffies(1000));
cxl: Support the cxl kernel API from a guest Like on bare-metal, the cxl driver creates a virtual PHB and a pci device for the AFU. The configuration space of the device is mapped to the configuration record of the AFU. Reuse the code defined in afu_cr_read8|16|32() when reading the configuration space of the AFU device. Even though the (virtual) AFU device is a pci device, the adapter is not. So a driver using the cxl kernel API cannot read the VPD of the adapter through the usual PCI interface. Therefore, we add a call to the cxl kernel API: ssize_t cxl_read_adapter_vpd(struct pci_dev *dev, void *buf, size_t count); Co-authored-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:40 +08:00
if ((rc = cxl_pci_vphb_add(afu)))
dev_info(&afu->dev, "Can't register vPHB\n");
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
return 0;
err_put2:
cxl_sysfs_afu_remove(afu);
err_put1:
device_unregister(&afu->dev);
free = false;
guest_release_serr_irq(afu);
err2:
guest_unmap_slice_regs(afu);
err1:
if (free) {
kfree(afu->guest);
kfree(afu);
}
return rc;
}
void cxl_guest_remove_afu(struct cxl_afu *afu)
{
if (!afu)
return;
cxl: Check periodically the coherent platform function's state In the PowerVM environment, the PHYP CoherentAccel component manages the state of the Coherent Accelerator Processor Interface adapter and virtualizes CAPI resources, handles CAPP, PSL, PSL Slice errors - and interrupts - and provides a new set of hcalls for the OS APIs to utilize Accelerator Function Unit (AFU). During the course of operation, a coherent platform function can encounter errors. Some possible reason for errors are: • Hardware recoverable and unrecoverable errors • Transient and over-threshold correctable errors PHYP implements its own state model for the coherent platform function. The state of the AFU is available through a hcall. The current implementation of the cxl driver, for the PowerVM environment, checks this state of the AFU only when an action is requested - open a device, ioctl command, memory map, attach/detach a process - from an external driver - cxlflash, libcxl. If an error is detected the cxl driver handles the error according the content of the Power Architecture Platform Requirements document. But in case of low-level troubles (or error injection), the PHYP component may reset the card and change the AFU state. The PHYP interface doesn't provide any way to be notified when that happens thus implies that the cxl driver: • cannot handle immediatly the state change of the AFU. • cannot notify other drivers (cxlflash, ...) The purpose of this patch is to wake up the cpu periodically to check the current state of each AFU and to see if we need to enter an error recovery path. Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-04-22 21:39:22 +08:00
/* flush and stop pending job */
afu->guest->handle_err = false;
flush_delayed_work(&afu->guest->work_err);
cxl: Support the cxl kernel API from a guest Like on bare-metal, the cxl driver creates a virtual PHB and a pci device for the AFU. The configuration space of the device is mapped to the configuration record of the AFU. Reuse the code defined in afu_cr_read8|16|32() when reading the configuration space of the AFU device. Even though the (virtual) AFU device is a pci device, the adapter is not. So a driver using the cxl kernel API cannot read the VPD of the adapter through the usual PCI interface. Therefore, we add a call to the cxl kernel API: ssize_t cxl_read_adapter_vpd(struct pci_dev *dev, void *buf, size_t count); Co-authored-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:40 +08:00
cxl_pci_vphb_remove(afu);
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
cxl_sysfs_afu_remove(afu);
spin_lock(&afu->adapter->afu_list_lock);
afu->adapter->afu[afu->slice] = NULL;
spin_unlock(&afu->adapter->afu_list_lock);
cxl_context_detach_all(afu);
cxl_ops->afu_deactivate_mode(afu, afu->current_mode);
guest_release_serr_irq(afu);
guest_unmap_slice_regs(afu);
device_unregister(&afu->dev);
}
static void free_adapter(struct cxl *adapter)
{
struct irq_avail *cur;
int i;
cxl: fix potential NULL dereference in free_adapter() If kzalloc() fails when allocating adapter->guest in cxl_guest_init_adapter(), we call free_adapter() before erroring out. free_adapter() in turn attempts to dereference adapter->guest, which in this case is NULL. In free_adapter(), skip the adapter->guest cleanup if adapter->guest is NULL. Fixes: 14baf4d9c739 ("cxl: Add guest-specific code") Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-07-15 15:20:36 +08:00
if (adapter->guest) {
if (adapter->guest->irq_avail) {
for (i = 0; i < adapter->guest->irq_nranges; i++) {
cur = &adapter->guest->irq_avail[i];
kfree(cur->bitmap);
}
kfree(adapter->guest->irq_avail);
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
}
cxl: fix potential NULL dereference in free_adapter() If kzalloc() fails when allocating adapter->guest in cxl_guest_init_adapter(), we call free_adapter() before erroring out. free_adapter() in turn attempts to dereference adapter->guest, which in this case is NULL. In free_adapter(), skip the adapter->guest cleanup if adapter->guest is NULL. Fixes: 14baf4d9c739 ("cxl: Add guest-specific code") Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-07-15 15:20:36 +08:00
kfree(adapter->guest->status);
kfree(adapter->guest);
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
}
cxl_remove_adapter_nr(adapter);
kfree(adapter);
}
static int properties_look_ok(struct cxl *adapter)
{
/* The absence of this property means that the operational
* status is unknown or okay
*/
if (strlen(adapter->guest->status) &&
strcmp(adapter->guest->status, "okay")) {
pr_err("ABORTING:Bad operational status of the device\n");
return -EINVAL;
}
return 0;
}
ssize_t cxl_guest_read_adapter_vpd(struct cxl *adapter, void *buf, size_t len)
{
return guest_collect_vpd(adapter, NULL, buf, len);
}
void cxl_guest_remove_adapter(struct cxl *adapter)
{
pr_devel("in %s\n", __func__);
cxl_sysfs_adapter_remove(adapter);
cxl: Support to flash a new image on the adapter from a guest The new flash.c file contains the logic to flash a new image on the adapter, through a hcall. It is an iterative process, with chunks of data of 1M at a time. There are also 2 phases: write and verify. The flash operation itself is driven from a user-land tool. Once flashing is successful, an rtas call is made to update the device tree with the new properties values for the adapter and the AFU(s) Add a new char device for the adapter, so that the flash tool can access the card, even if there is no valid AFU on it. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:38 +08:00
cxl_guest_remove_chardev(adapter);
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
device_unregister(&adapter->dev);
}
static void release_adapter(struct device *dev)
{
free_adapter(to_cxl_adapter(dev));
}
struct cxl *cxl_guest_init_adapter(struct device_node *np, struct platform_device *pdev)
{
struct cxl *adapter;
bool free = true;
int rc;
if (!(adapter = cxl_alloc_adapter()))
return ERR_PTR(-ENOMEM);
if (!(adapter->guest = kzalloc(sizeof(struct cxl_guest), GFP_KERNEL))) {
free_adapter(adapter);
return ERR_PTR(-ENOMEM);
}
adapter->slices = 0;
adapter->guest->pdev = pdev;
adapter->dev.parent = &pdev->dev;
adapter->dev.release = release_adapter;
dev_set_drvdata(&pdev->dev, adapter);
cxl: Allow initialization on timebase sync failures Failure to synchronize the PSL timebase currently prevents the initialization of the cxl card, thus rendering the card useless. This is too extreme for a feature which is rarely used, if at all. No hardware AFUs or software is currently using PSL timebase. This patch still tries to synchronize the PSL timebase when the card is initialized, but ignores the error if it can't. Instead, it reports a status via /sys. Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-22 03:32:48 +08:00
/*
* Hypervisor controls PSL timebase initialization (p1 register).
* On FW840, PSL is initialized.
*/
adapter->psl_timebase_synced = true;
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
if ((rc = cxl_of_read_adapter_handle(adapter, np)))
goto err1;
if ((rc = cxl_of_read_adapter_properties(adapter, np)))
goto err1;
if ((rc = properties_look_ok(adapter)))
goto err1;
cxl: Support to flash a new image on the adapter from a guest The new flash.c file contains the logic to flash a new image on the adapter, through a hcall. It is an iterative process, with chunks of data of 1M at a time. There are also 2 phases: write and verify. The flash operation itself is driven from a user-land tool. Once flashing is successful, an rtas call is made to update the device tree with the new properties values for the adapter and the AFU(s) Add a new char device for the adapter, so that the flash tool can access the card, even if there is no valid AFU on it. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:38 +08:00
if ((rc = cxl_guest_add_chardev(adapter)))
goto err1;
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
/*
* After we call this function we must not free the adapter directly,
* even if it returns an error!
*/
if ((rc = cxl_register_adapter(adapter)))
goto err_put1;
if ((rc = cxl_sysfs_adapter_add(adapter)))
goto err_put1;
cxl: Prevent adapter reset if an active context exists This patch prevents resetting the cxl adapter via sysfs in presence of one or more active cxl_context on it. This protects against an unrecoverable error caused by PSL owning a dirty cache line even after reset and host tries to touch the same cache line. In case a force reset of the card is required irrespective of any active contexts, the int value -1 can be stored in the 'reset' sysfs attribute of the card. The patch introduces a new atomic_t member named contexts_num inside struct cxl that holds the number of active context attached to the card , which is checked against '0' before proceeding with the reset. To prevent against a race condition where a context is activated just after reset check is performed, the contexts_num is atomically set to '-1' after reset-check to indicate that no more contexts can be activated on the card anymore. Before activating a context we atomically test if contexts_num is non-negative and if so, increment its value by one. In case the value of contexts_num is negative then it indicates that the card is about to be reset and context activation is error-ed out at that point. Fixes: 62fa19d4b4fd ("cxl: Add ability to reset the card") Cc: stable@vger.kernel.org # v4.0+ Acked-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Reviewed-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com> Signed-off-by: Vaibhav Jain <vaibhav@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-10-14 17:38:36 +08:00
/* release the context lock as the adapter is configured */
cxl_adapter_context_unlock(adapter);
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
return adapter;
err_put1:
device_unregister(&adapter->dev);
free = false;
cxl: Support to flash a new image on the adapter from a guest The new flash.c file contains the logic to flash a new image on the adapter, through a hcall. It is an iterative process, with chunks of data of 1M at a time. There are also 2 phases: write and verify. The flash operation itself is driven from a user-land tool. Once flashing is successful, an rtas call is made to update the device tree with the new properties values for the adapter and the AFU(s) Add a new char device for the adapter, so that the flash tool can access the card, even if there is no valid AFU on it. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:38 +08:00
cxl_guest_remove_chardev(adapter);
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
err1:
if (free)
free_adapter(adapter);
return ERR_PTR(rc);
}
cxl: Support to flash a new image on the adapter from a guest The new flash.c file contains the logic to flash a new image on the adapter, through a hcall. It is an iterative process, with chunks of data of 1M at a time. There are also 2 phases: write and verify. The flash operation itself is driven from a user-land tool. Once flashing is successful, an rtas call is made to update the device tree with the new properties values for the adapter and the AFU(s) Add a new char device for the adapter, so that the flash tool can access the card, even if there is no valid AFU on it. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:38 +08:00
void cxl_guest_reload_module(struct cxl *adapter)
{
struct platform_device *pdev;
pdev = adapter->guest->pdev;
cxl_guest_remove_adapter(adapter);
cxl_of_probe(pdev);
}
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
const struct cxl_backend_ops cxl_guest_ops = {
.module = THIS_MODULE,
.adapter_reset = guest_reset,
.alloc_one_irq = guest_alloc_one_irq,
.release_one_irq = guest_release_one_irq,
.alloc_irq_ranges = guest_alloc_irq_ranges,
.release_irq_ranges = guest_release_irq_ranges,
.setup_irq = NULL,
.handle_psl_slice_error = guest_handle_psl_slice_error,
.psl_interrupt = guest_psl_irq,
.ack_irq = guest_ack_irq,
.attach_process = guest_attach_process,
.detach_process = guest_detach_process,
cxl: Update process element after allocating interrupts In the kernel API, it is possible to attempt to allocate AFU interrupts after already starting a context. Since the process element structure used by the hardware is only filled out at the time the context is started, it will not be updated with the interrupt numbers that have just been allocated and therefore AFU interrupts will not work unless they were allocated prior to starting the context. This can present some difficulties as each CAPI enabled PCI device in the kernel API has a default context, which may need to be started very early to enable translations, potentially before interrupts can easily be set up. This patch makes the API more flexible to allow interrupts to be allocated after a context has already been started and takes care of updating the PE structure used by the hardware and notifying it to discard any cached copy it may have. The update is currently performed via a terminate/remove/add sequence. This is necessary on some hardware such as the XSL that does not properly support the update LLCMD. Note that this is only supported on powernv at present - attempting to perform this ordering on PowerVM will raise a warning. Signed-off-by: Ian Munsie <imunsie@au1.ibm.com> Reviewed-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-05-24 00:14:05 +08:00
.update_ivtes = NULL,
cxl: sysfs support for guests Filter out a few adapter parameters which don't make sense in a guest. Document the changes. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:37 +08:00
.support_attributes = guest_support_attributes,
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
.link_ok = guest_link_ok,
.release_afu = guest_release_afu,
.afu_read_err_buffer = guest_afu_read_err_buffer,
.afu_check_and_enable = guest_afu_check_and_enable,
.afu_activate_mode = guest_afu_activate_mode,
.afu_deactivate_mode = guest_afu_deactivate_mode,
.afu_reset = guest_afu_reset,
.afu_cr_read8 = guest_afu_cr_read8,
.afu_cr_read16 = guest_afu_cr_read16,
.afu_cr_read32 = guest_afu_cr_read32,
.afu_cr_read64 = guest_afu_cr_read64,
cxl: Support the cxl kernel API from a guest Like on bare-metal, the cxl driver creates a virtual PHB and a pci device for the AFU. The configuration space of the device is mapped to the configuration record of the AFU. Reuse the code defined in afu_cr_read8|16|32() when reading the configuration space of the AFU device. Even though the (virtual) AFU device is a pci device, the adapter is not. So a driver using the cxl kernel API cannot read the VPD of the adapter through the usual PCI interface. Therefore, we add a call to the cxl kernel API: ssize_t cxl_read_adapter_vpd(struct pci_dev *dev, void *buf, size_t count); Co-authored-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:40 +08:00
.afu_cr_write8 = guest_afu_cr_write8,
.afu_cr_write16 = guest_afu_cr_write16,
.afu_cr_write32 = guest_afu_cr_write32,
.read_adapter_vpd = cxl_guest_read_adapter_vpd,
cxl: Add guest-specific code The new of.c file contains code to parse the device tree to find out about cxl adapters and AFUs. guest.c implements the guest-specific callbacks for the backend API. The process element ID is not known until the context is attached, so we have to separate the context ID assigned by the cxl driver from the process element ID visible to the user applications. In bare-metal, the 2 IDs match. Co-authored-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by: Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by: Manoj Kumar <manoj@linux.vnet.ibm.com> Acked-by: Ian Munsie <imunsie@au1.ibm.com> [mpe: Fix SMP=n build, fix PSERIES=n build, minor whitespace fixes] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-03-04 19:26:36 +08:00
};