From: Christophe Lombard <redacted>
The AFU Information DVSEC capability is a means to extract common,
general information about all of the AFUs associated with a Function
independent of the specific functionality that each AFU provides.
This patch fixes the access to the AFU Descriptor Data indexed by the
AFU Info Index field.
Fixes: 5ef3166e8a32 ("ocxl: Driver code for 'generic' opencapi devices")
Signed-off-by: Christophe Lombard <redacted>
---
drivers/misc/ocxl/config.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/misc/ocxl/config.c b/drivers/misc/ocxl/config.c
index 2e30de9..de01623 100644
--- a/drivers/misc/ocxl/config.c
+++ b/drivers/misc/ocxl/config.c
@@ -280,7 +280,9 @@ int ocxl_config_check_afu_index(struct pci_dev *dev,
u32 val;
int rc, templ_major, templ_minor, len;
- pci_write_config_word(dev, fn->dvsec_afu_info_pos, afu_idx);
+ pci_write_config_word(dev,
+ fn->dvsec_afu_info_pos + OCXL_DVSEC_AFU_INFO_AFU_IDX,
+ afu_idx);
rc = read_afu_info(dev, fn, OCXL_DVSEC_TEMPL_VERSION, &val);
if (rc)
return rc;
--
2.7.4
On 13/08/18 19:01, Christophe Lombard wrote:
From: Christophe Lombard <redacted>
Your git committer email should probably match your sign-off email.
The AFU Information DVSEC capability is a means to extract common,
general information about all of the AFUs associated with a Function
independent of the specific functionality that each AFU provides.
This patch fixes the access to the AFU Descriptor Data indexed by the
AFU Info Index field.
Fixes: 5ef3166e8a32 ("ocxl: Driver code for 'generic' opencapi devices")
Signed-off-by: Christophe Lombard <redacted>
This looks like it should go to stable? I assume the reason we haven't
noticed this previously is because we have not been testing with
multi-AFU cards.
quoted hunk
---
drivers/misc/ocxl/config.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/misc/ocxl/config.c b/drivers/misc/ocxl/config.c
index 2e30de9..de01623 100644
--- a/drivers/misc/ocxl/config.c
+++ b/drivers/misc/ocxl/config.c
@@ -280,7 +280,9 @@ int ocxl_config_check_afu_index(struct pci_dev *dev,
u32 val;
int rc, templ_major, templ_minor, len;
- pci_write_config_word(dev, fn->dvsec_afu_info_pos, afu_idx);
+ pci_write_config_word(dev,
+ fn->dvsec_afu_info_pos + OCXL_DVSEC_AFU_INFO_AFU_IDX,
+ afu_idx);
pci_write_config_byte() would be more appropriate here (see
ocxl_config_read_afu() at line 454)
rc = read_afu_info(dev, fn, OCXL_DVSEC_TEMPL_VERSION, &val);
if (rc)
return rc;
--
Andrew Donnellan OzLabs, ADL Canberra
andrew.donnellan@au1.ibm.com IBM Australia Limited
Le 13/08/2018 à 11:48, Andrew Donnellan a écrit :
On 13/08/18 19:01, Christophe Lombard wrote:
quoted
From: Christophe Lombard <redacted>
Your git committer email should probably match your sign-off email.
quoted
The AFU Information DVSEC capability is a means to extract common,
general information about all of the AFUs associated with a Function
independent of the specific functionality that each AFU provides.
This patch fixes the access to the AFU Descriptor Data indexed by the
AFU Info Index field.
Fixes: 5ef3166e8a32 ("ocxl: Driver code for 'generic' opencapi devices")
Signed-off-by: Christophe Lombard <redacted>
This looks like it should go to stable? I assume the reason we haven't
noticed this previously is because we have not been testing with
multi-AFU cards.
Yes, it's hidden until we have more than one AFU per function. Also, the
field we overwrite at the wrong offset is read-only, so it goes undetected.
Fred
quoted
---
drivers/misc/ocxl/config.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/misc/ocxl/config.c b/drivers/misc/ocxl/config.c
index 2e30de9..de01623 100644
--- a/drivers/misc/ocxl/config.c
+++ b/drivers/misc/ocxl/config.c
@@ -280,7 +280,9 @@ int ocxl_config_check_afu_index(struct pci_dev *dev,
u32 val;
int rc, templ_major, templ_minor, len;
- pci_write_config_word(dev, fn->dvsec_afu_info_pos, afu_idx);
+ pci_write_config_word(dev,
+ fn->dvsec_afu_info_pos + OCXL_DVSEC_AFU_INFO_AFU_IDX,
+ afu_idx);
pci_write_config_byte() would be more appropriate here (see
ocxl_config_read_afu() at line 454)
quoted
rc = read_afu_info(dev, fn, OCXL_DVSEC_TEMPL_VERSION, &val);
if (rc)
return rc;
Le 13/08/2018 à 11:48, Andrew Donnellan a écrit :
On 13/08/18 19:01, Christophe Lombard wrote:
quoted
From: Christophe Lombard <redacted>
Your git committer email should probably match your sign-off email.
quoted
The AFU Information DVSEC capability is a means to extract common,
general information about all of the AFUs associated with a Function
independent of the specific functionality that each AFU provides.
This patch fixes the access to the AFU Descriptor Data indexed by the
AFU Info Index field.
Fixes: 5ef3166e8a32 ("ocxl: Driver code for 'generic' opencapi devices")
Signed-off-by: Christophe Lombard <redacted>
This looks like it should go to stable? I assume the reason we haven't
noticed this previously is because we have not been testing with
multi-AFU cards.
quoted
---
drivers/misc/ocxl/config.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/misc/ocxl/config.c b/drivers/misc/ocxl/config.c
index 2e30de9..de01623 100644
--- a/drivers/misc/ocxl/config.c
+++ b/drivers/misc/ocxl/config.c
@@ -280,7 +280,9 @@ int ocxl_config_check_afu_index(struct pci_dev *dev,
u32 val;
int rc, templ_major, templ_minor, len;
- pci_write_config_word(dev, fn->dvsec_afu_info_pos, afu_idx);
+ pci_write_config_word(dev,
+ fn->dvsec_afu_info_pos + OCXL_DVSEC_AFU_INFO_AFU_IDX,
+ afu_idx);
pci_write_config_byte() would be more appropriate here (see
ocxl_config_read_afu() at line 454)
quoted
rc = read_afu_info(dev, fn, OCXL_DVSEC_TEMPL_VERSION, &val);
if (rc)
return rc;
right. Thanks for the review.