linux/drivers/edac, branch v5.0

EDAC, altera: Fix S10 persistent register offset

2019-01-24T16:13:59Z

Correct the persistent register offset where address and status are stored. Fixes: 08f08bfb7b4c ("EDAC, altera: Merge Stratix10 into the Arria10 SDRAM probe routine") Signed-off-by: Thor Thayer Signed-off-by: Borislav Petkov Cc: James Morse Cc: Mauro Carvalho Chehab Cc: devicetree@vger.kernel.org Cc: dinguyen@kernel.org Cc: linux-edac Cc: mark.rutland@arm.com Cc: robh+dt@kernel.org Cc: stable Link: https://lkml.kernel.org/r/1548179287-21760-2-git-send-email-thor.thayer@linux.intel.com

EDAC, fsl_ddr: Add LS1021A to the list of supported hardware

2018-12-19T10:57:45Z

The Freescale ddr driver also works on the LS1021A board. Signed-off-by: Patrick Havelange Signed-off-by: Borislav Petkov Cc: Mauro Carvalho Chehab Cc: York Sun Cc: arnout.vandecappelle@essensium.com Cc: linux-edac Cc: matthew.weber@rockwellcollins.com Cc: patrick.havelange@essensium.com Link: https://lkml.kernel.org/r/20181219104323.10324-1-patrick.havelange@essensium.com

EDAC, i5000: Remove set but not used local variables

2018-12-11T13:53:49Z

Remove unused local variables as reported by gcc's -Wunused-but-set-variable option. [ bp: simplify commit message. ] Signed-off-by: YueHaibing Signed-off-by: Borislav Petkov Cc: Mauro Carvalho Chehab Cc: linux-edac Link: https://lkml.kernel.org/r/20181211095207.25936-1-yuehaibing@huawei.com

EDAC, i82975x: Fix spelling mistake "reserverd" -> "reserved"

2018-11-20T16:46:01Z

Fix a spelling mistake in a register layout description. Signed-off-by: Alexandre Belloni Signed-off-by: Borislav Petkov Cc: "Arvind R." Cc: linux-edac Link: https://lkml.kernel.org/r/20181120153304.1218-1-alexandre.belloni@bootlin.com

EDAC, fsl: Move error injection under CONFIG_EDAC_DEBUG

2018-11-20T16:33:32Z

Gate error injection feature with CONFIG_EDAC_DEBUG so that it is not visible in production setups. Suggested-by: Borislav Petkov Signed-off-by: York Sun Signed-off-by: Borislav Petkov Cc: linux-edac Link: https://lkml.kernel.org/r/20181119225303.13265-1-york.sun@nxp.com

EDAC, skx: Let EDAC core show the decoded result for debugfs

2018-11-16T10:19:53Z

Current debugfs shows the decoded result in its own print format which is inconvenient for analysis/statistics. Use skx_mce_check_error() instead of skx_decode() for debugfs, then the decoded result is showed via EDAC core in a more readable format like "CPU_SrcID#[0-9]_MC#[0-9]_Chan#[0-9]_DIMM#[0-9]". Print a warning the first time this interface is used so the administrator can see the console log that error(s) have been faked. Co-developed-by: Tony Luck Signed-off-by: Qiuxu Zhuo Signed-off-by: Borislav Petkov CC: Mauro Carvalho Chehab CC: arozansk@redhat.com CC: linux-edac Link: http://lkml.kernel.org/r/1542353705-13531-1-git-send-email-qiuxu.zhuo@intel.com

EDAC, skx: Move debugfs node under EDAC's hierarchy

2018-11-16T10:17:24Z

The debugfs node is /sys/kernel/debug/skx_edac_test. Rename it and move under EDAC debugfs root directory. Remove the unused 'skx_fake_addr' and remove the 'skx_test' on error. Co-developed-by: Tony Luck Signed-off-by: Qiuxu Zhuo Signed-off-by: Borislav Petkov CC: Mauro Carvalho Chehab CC: arozansk@redhat.com CC: linux-edac Link: http://lkml.kernel.org/r/1542353684-13496-1-git-send-email-qiuxu.zhuo@intel.com

EDAC, skx: Prepend hex formatting with '0x'

2018-11-16T10:17:19Z

Some debug/error strings in hex formatting do not have the '0x' prefix. Prepend hex formatting with '0x' for them, but with one exception: "Couldn't enable %04x:%04x", instead of putting '0x' in this line, add the word 'device'. We commonly use 8086:1234 without the leading '0x' (e.g. as '-d' argument to lspci(8) and setpci(8) commands). Suggested-by: Borislav Petkov Signed-off-by: Qiuxu Zhuo Signed-off-by: Borislav Petkov CC: Mauro Carvalho Chehab CC: Tony Luck CC: arozansk@redhat.com CC: linux-edac Link: http://lkml.kernel.org/r/1542353660-13458-1-git-send-email-qiuxu.zhuo@intel.com

EDAC, skx: Fix function calling order in skx_exit()

2018-11-16T10:17:11Z

The order of function calling in skx_exit() is not the reversed order in skx_init(). Fix it. Signed-off-by: Qiuxu Zhuo Signed-off-by: Borislav Petkov CC: Mauro Carvalho Chehab CC: Tony Luck CC: arozansk@redhat.com CC: linux-edac Link: http://lkml.kernel.org/r/1542353616-13421-1-git-send-email-qiuxu.zhuo@intel.com

EDAC: Drop per-memory controller buses

2018-11-13T20:55:24Z

... and use the single edac_subsys object returned from subsys_system_register(). The idea is to have a single bus and multiple devices on it. Signed-off-by: Borislav Petkov Acked-by: Mauro Carvalho Chehab CC: Aristeu Rozanski Filho CC: Greg KH CC: Justin Ernst CC: linux-edac CC: Mauro Carvalho Chehab CC: Russ Anderson Cc: Tony Luck Link: https://lkml.kernel.org/r/20180926152752.GG5584@zn.tnic