Age | Commit message (Collapse) | Author |
|
Add a new Intel Alder Lake-N SoC compute die ID for EDAC support.
Signed-off-by: Lili Li <lili.li@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Qiuxu Zhuo <qiuxu.zhuo@intel.com>
Link: https://lore.kernel.org/r/20240129062040.60809-2-qiuxu.zhuo@intel.com
|
|
Add Intel Meteor Lake-P SoC compute die IDs for EDAC support.
These Meteor Lake-P SoCs share similar IBECC registers with
Alder Lake-P SoCs.
Signed-off-by: Qiuxu Zhuo <qiuxu.zhuo@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
|
|
Add Intel Meteor Lake-PS SoC compute die IDs for EDAC support.
These SoCs share similar IBECC registers with Alder Lake-P SoCs.
The only difference is that IBECC presence is detected through an
MMIO-mapped register instead of the capability register in the
PCI configuration space.
Signed-off-by: Qiuxu Zhuo <qiuxu.zhuo@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
|
|
Add Intel Raptor Lake-P SoC compute die IDs for EDAC support.
These Raptor Lake-P SoCs share similar IBECC registers with Alder Lake-P
SoCs but extend the most significant bit of the error address logged
in IBECC from bit 38 to bit 45.
Signed-off-by: Qiuxu Zhuo <qiuxu.zhuo@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
|
|
Add Intel Alder Lake-N SoC compute die IDs for EDAC support.
Alder Lake-N, with one memory controller, is a reduced version of
Alder Lake-P, which has two memory controllers.
Signed-off-by: Qiuxu Zhuo <qiuxu.zhuo@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
|
|
Make get_mchbar() helper function to retrieve the BAR address of
the memory controller. No function changes.
Signed-off-by: Qiuxu Zhuo <qiuxu.zhuo@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
|
|
Current igen6_edac checks for pending errors before the registration
of the error handler. However, there is a possibility that the error
occurs during the registration process, leading to unhandled pending
errors and no future error events. This issue can be reproduced by
repeatedly injecting errors during the loading of the igen6_edac.
Fix this issue by moving the pending error handler after the registration
of the error handler, ensuring that no pending errors are left unhandled.
Fixes: 10590a9d4f23 ("EDAC/igen6: Add EDAC driver for Intel client SoCs using IBECC")
Reported-by: Ee Wey Lim <ee.wey.lim@intel.com>
Tested-by: Ee Wey Lim <ee.wey.lim@intel.com>
Signed-off-by: Qiuxu Zhuo <qiuxu.zhuo@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Link: https://lore.kernel.org/r/20230725080427.23883-1-qiuxu.zhuo@intel.com
|
|
Return -EBUSY instead of -ENODEV just like the other EDAC drivers do.
[ bp: Rewrite text. ]
Signed-off-by: Jia He <justin.he@arm.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lore.kernel.org/r/20221018082214.569504-8-justin.he@arm.com
|
|
Call ghes_get_devices() to check whether ghes_edac should be used on the
platform where it is preferred over the corresponding chipset-specific
EDAC driver.
Unlike the existing edac_get_owner() check, the ghes_get_devices() check
works independent to the module_init ordering.
[ bp: Massage. ]
Suggested-by: Toshi Kani <toshi.kani@hpe.com>
Signed-off-by: Jia He <justin.he@arm.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lore.kernel.org/r/20221010023559.69655-6-justin.he@arm.com
|
|
Alder Lake SoC shares the same memory controller and In-Band ECC
(IBECC) IP with Tiger Lake SoC. Like Tiger Lake, it also has two
memory controllers each associated one IBECC instance. The minor
differences include the MMIO offset of each memory controller and
the type of memory error address logged in the IBECC.
So add Alder Lake compute die IDs, adjust the MMIO offset for each
memory controller and handle the type of memory error address logged
in the IBECC for Alder Lake EDAC support.
Tested-by: Vrukesh V Panse <vrukesh.v.panse@intel.com>
Signed-off-by: Qiuxu Zhuo <qiuxu.zhuo@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Link: https://lore.kernel.org/r/20210611170123.1057025-7-tony.luck@intel.com
|
|
Tiger Lake SoC shares the same memory controller and In-Band ECC
(IBECC) IP with Elkhart Lake SoC. The main differences are that Tiger
Lake has two memory controllers each associated with one IBECC and
uses Machine Check for the memory error notification.
So add Tiger Lake compute die IDs, MCE decoding chain registration,
and memory slice decoding for Tiger Lake EDAC support.
Signed-off-by: Qiuxu Zhuo <qiuxu.zhuo@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Link: https://lore.kernel.org/r/20210611170123.1057025-6-tony.luck@intel.com
|
|
The Ice Lake Neural Network Processor for Deep Learning Inference
(ICL-NNPI) SoC shares the same memory controller and In-Band ECC with
Elkhart Lake SoC. Add the ICL-NNPI compute die IDs for EDAC support.
Signed-off-by: Qiuxu Zhuo <qiuxu.zhuo@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Link: https://lore.kernel.org/r/20210611170123.1057025-5-tony.luck@intel.com
|
|
Fixes: 10590a9d4f23 ("EDAC/igen6: Add EDAC driver for Intel client SoCs using IBECC")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/r/20201123031850.GA20416@aef56166e5fc
Signed-off-by: Tony Luck <tony.luck@intel.com>
|
|
Add debugfs support to fake memory correctable errors to test the
error reporting path and the error address decoding logic in the
igen6_edac driver.
Please note that the fake errors are also reported to EDAC core and
then the CE counter in EDAC sysfs is also increased.
Signed-off-by: Qiuxu Zhuo <qiuxu.zhuo@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
|
|
This driver supports Intel client SoC with integrated memory controller
using In-Band ECC(IBECC). The memory correctable and uncorrectable errors
are reported via NMIs. The driver handles the NMIs and decodes the memory
error address to platform specific address. The first IBECC-supported SoC
is Elkhart Lake.
[Tony: s/#include <linux/nmi.h>/#include <asm/nmi.h>/ to fix randconfig build]
Signed-off-by: Qiuxu Zhuo <qiuxu.zhuo@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
|