| Message ID | cover.1681838291.git.alison.schofield@intel.com |
|---|---|
| Headers | show |
| Series | CXL Poison List Retrieval & Tracing | expand |
On Tue, 18 Apr 2023 10:39:00 -0700 alison.schofield@intel.com wrote: > From: Alison Schofield <alison.schofield@intel.com> > FWIW I've just been hammering the QEMU emulation for this to test a new version of that, but as a side effect I hit ther corner cases with this as well and it all looks good to me. Tested-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> > Changes in v13: > - New Lead-in patches > cxl/mbox: Deprecate poison commands (Dan) > cxl/mbox: Restrict poison cmds to debugfs cxl_raw_allow_all > > - New Patch: cxl/mbox: Initialize the poison state > Patch connects the lead-in patches with the rest of this set. Poison init > was previously done in the GET_POISON_LIST patch. With LIST deprecated, > needed a method, along with a reason, to discover device support. > > - cxl_poison_state_init(): use kvmalloc for potentially large payload (Dan) > - cxl_poison_state_init() unset poison enabled bit on failure > - trigger sysfs: make the core interface a proper api (Dan) > - trigger sysfs: use down_read_interruptible (Dan) > - Reorganize the by_endpoint work to make typesafe (Dan) > - poison_by_decoder() only fill ctx when iteration is done > - Remove mentions of mixed mode as a 'watch for'. Just say no. (Dan) > - s/overflow_t/overflow_ts in cxlmem.h struct and trace.h struct (Dan) > - Really remove errant line from cxl_memdev_visible() (Jonathan, DaveJ, Dan) > > Link to v12: > https://lore.kernel.org/linux-cxl/cover.1681159309.git.alison.schofield@intel.com/ > > Add support for retrieving device poison lists and store the returned > error records as kernel trace events. > > The handling of the poison list is guided by the CXL 3.0 Specification > Section 8.2.9.8.4.1. [1] > > Example trigger: > $ echo 1 > /sys/bus/cxl/devices/mem0/trigger_poison_list > > Example Trace Events: > > Poison found in a PMEM Region: > cxl_poison: memdev=mem0 host=cxl_mem.0 serial=0 trace_type=List region=region11 region_uuid=d96e67ec-76b0-406f-8c35-5b52630dcad1 hpa=0xf100000000 dpa=0x70000000 dpa_length=0x40 source=Injected flags= overflow_time=0 > > Poison found in RAM Region: > cxl_poison: memdev=mem0 host=cxl_mem.0 serial=0 trace_type=List region=region2 region_uuid=00000000-0000-0000-0000-000000000000 hpa=0xf010000000 dpa=0x0 dpa_length=0x40 source=Injected flags= overflow_time=0 > > Poison found in an unmapped DPA resource: > cxl_poison: memdev=mem3 host=cxl_mem.3 serial=3 trace_type=List region= region_uuid=00000000-0000-0000-0000-000000000000 hpa=0xffffffffffffffff dpa=0x40000000 dpa_length=0x40 source=Injected flags= overflow_time=0 > > [1]: https://www.computeexpresslink.org/download-the-specification > > Alison Schofield (8): > cxl/mbox: Restrict poison cmds to debugfs cxl_raw_allow_all > cxl/mbox: Initialize the poison state > cxl/mbox: Add GET_POISON_LIST mailbox command > cxl/trace: Add TRACE support for CXL media-error records > cxl/memdev: Add trigger_poison_list sysfs attribute > cxl/region: Provide region info to the cxl_poison trace event > cxl/trace: Add an HPA to cxl_poison trace events > tools/testing/cxl: Mock support for Get Poison List > > Dan Williams (1): > cxl/mbox: Deprecate poison commands > > Documentation/ABI/testing/sysfs-bus-cxl | 14 +++ > drivers/cxl/core/core.h | 9 ++ > drivers/cxl/core/mbox.c | 150 ++++++++++++++++++++++-- > drivers/cxl/core/memdev.c | 54 +++++++++ > drivers/cxl/core/region.c | 124 ++++++++++++++++++++ > drivers/cxl/core/trace.c | 94 +++++++++++++++ > drivers/cxl/core/trace.h | 101 ++++++++++++++++ > drivers/cxl/cxlmem.h | 83 ++++++++++++- > drivers/cxl/mem.c | 43 +++++++ > drivers/cxl/pci.c | 4 + > include/uapi/linux/cxl_mem.h | 35 +++++- > tools/testing/cxl/test/mem.c | 42 +++++++ > 12 files changed, 740 insertions(+), 13 deletions(-) > > > base-commit: e686c32590f40bffc45f105c04c836ffad3e531a
On Sun, 23 Apr 2023 16:30:11 +0100 Jonathan Cameron <Jonathan.Cameron@huawei.com> wrote: > On Tue, 18 Apr 2023 10:39:00 -0700 > alison.schofield@intel.com wrote: > > > From: Alison Schofield <alison.schofield@intel.com> > > > > FWIW I've just been hammering the QEMU emulation for this to test a new > version of that, but as a side effect I hit ther corner cases with this as well and > it all looks good to me. > > Tested-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> I should refine that slightly - doesn't cover patch 9 as I didn't try the mocking. Jonathan > > > > Changes in v13: > > - New Lead-in patches > > cxl/mbox: Deprecate poison commands (Dan) > > cxl/mbox: Restrict poison cmds to debugfs cxl_raw_allow_all > > > > - New Patch: cxl/mbox: Initialize the poison state > > Patch connects the lead-in patches with the rest of this set. Poison init > > was previously done in the GET_POISON_LIST patch. With LIST deprecated, > > needed a method, along with a reason, to discover device support. > > > > - cxl_poison_state_init(): use kvmalloc for potentially large payload (Dan) > > - cxl_poison_state_init() unset poison enabled bit on failure > > - trigger sysfs: make the core interface a proper api (Dan) > > - trigger sysfs: use down_read_interruptible (Dan) > > - Reorganize the by_endpoint work to make typesafe (Dan) > > - poison_by_decoder() only fill ctx when iteration is done > > - Remove mentions of mixed mode as a 'watch for'. Just say no. (Dan) > > - s/overflow_t/overflow_ts in cxlmem.h struct and trace.h struct (Dan) > > - Really remove errant line from cxl_memdev_visible() (Jonathan, DaveJ, Dan) > > > > Link to v12: > > https://lore.kernel.org/linux-cxl/cover.1681159309.git.alison.schofield@intel.com/ > > > > Add support for retrieving device poison lists and store the returned > > error records as kernel trace events. > > > > The handling of the poison list is guided by the CXL 3.0 Specification > > Section 8.2.9.8.4.1. [1] > > > > Example trigger: > > $ echo 1 > /sys/bus/cxl/devices/mem0/trigger_poison_list > > > > Example Trace Events: > > > > Poison found in a PMEM Region: > > cxl_poison: memdev=mem0 host=cxl_mem.0 serial=0 trace_type=List region=region11 region_uuid=d96e67ec-76b0-406f-8c35-5b52630dcad1 hpa=0xf100000000 dpa=0x70000000 dpa_length=0x40 source=Injected flags= overflow_time=0 > > > > Poison found in RAM Region: > > cxl_poison: memdev=mem0 host=cxl_mem.0 serial=0 trace_type=List region=region2 region_uuid=00000000-0000-0000-0000-000000000000 hpa=0xf010000000 dpa=0x0 dpa_length=0x40 source=Injected flags= overflow_time=0 > > > > Poison found in an unmapped DPA resource: > > cxl_poison: memdev=mem3 host=cxl_mem.3 serial=3 trace_type=List region= region_uuid=00000000-0000-0000-0000-000000000000 hpa=0xffffffffffffffff dpa=0x40000000 dpa_length=0x40 source=Injected flags= overflow_time=0 > > > > [1]: https://www.computeexpresslink.org/download-the-specification > > > > Alison Schofield (8): > > cxl/mbox: Restrict poison cmds to debugfs cxl_raw_allow_all > > cxl/mbox: Initialize the poison state > > cxl/mbox: Add GET_POISON_LIST mailbox command > > cxl/trace: Add TRACE support for CXL media-error records > > cxl/memdev: Add trigger_poison_list sysfs attribute > > cxl/region: Provide region info to the cxl_poison trace event > > cxl/trace: Add an HPA to cxl_poison trace events > > tools/testing/cxl: Mock support for Get Poison List > > > > Dan Williams (1): > > cxl/mbox: Deprecate poison commands > > > > Documentation/ABI/testing/sysfs-bus-cxl | 14 +++ > > drivers/cxl/core/core.h | 9 ++ > > drivers/cxl/core/mbox.c | 150 ++++++++++++++++++++++-- > > drivers/cxl/core/memdev.c | 54 +++++++++ > > drivers/cxl/core/region.c | 124 ++++++++++++++++++++ > > drivers/cxl/core/trace.c | 94 +++++++++++++++ > > drivers/cxl/core/trace.h | 101 ++++++++++++++++ > > drivers/cxl/cxlmem.h | 83 ++++++++++++- > > drivers/cxl/mem.c | 43 +++++++ > > drivers/cxl/pci.c | 4 + > > include/uapi/linux/cxl_mem.h | 35 +++++- > > tools/testing/cxl/test/mem.c | 42 +++++++ > > 12 files changed, 740 insertions(+), 13 deletions(-) > > > > > > base-commit: e686c32590f40bffc45f105c04c836ffad3e531a >
Jonathan Cameron wrote: > On Sun, 23 Apr 2023 16:30:11 +0100 > Jonathan Cameron <Jonathan.Cameron@huawei.com> wrote: > > > On Tue, 18 Apr 2023 10:39:00 -0700 > > alison.schofield@intel.com wrote: > > > > > From: Alison Schofield <alison.schofield@intel.com> > > > > > > > FWIW I've just been hammering the QEMU emulation for this to test a new > > version of that, but as a side effect I hit ther corner cases with this as well and > > it all looks good to me. > > > > Tested-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> > > I should refine that slightly - doesn't cover patch 9 as I didn't > try the mocking. Thanks Jonathan, I've picked up your test and review tags.
From: Alison Schofield <alison.schofield@intel.com> Changes in v13: - New Lead-in patches cxl/mbox: Deprecate poison commands (Dan) cxl/mbox: Restrict poison cmds to debugfs cxl_raw_allow_all - New Patch: cxl/mbox: Initialize the poison state Patch connects the lead-in patches with the rest of this set. Poison init was previously done in the GET_POISON_LIST patch. With LIST deprecated, needed a method, along with a reason, to discover device support. - cxl_poison_state_init(): use kvmalloc for potentially large payload (Dan) - cxl_poison_state_init() unset poison enabled bit on failure - trigger sysfs: make the core interface a proper api (Dan) - trigger sysfs: use down_read_interruptible (Dan) - Reorganize the by_endpoint work to make typesafe (Dan) - poison_by_decoder() only fill ctx when iteration is done - Remove mentions of mixed mode as a 'watch for'. Just say no. (Dan) - s/overflow_t/overflow_ts in cxlmem.h struct and trace.h struct (Dan) - Really remove errant line from cxl_memdev_visible() (Jonathan, DaveJ, Dan) Link to v12: https://lore.kernel.org/linux-cxl/cover.1681159309.git.alison.schofield@intel.com/ Add support for retrieving device poison lists and store the returned error records as kernel trace events. The handling of the poison list is guided by the CXL 3.0 Specification Section 8.2.9.8.4.1. [1] Example trigger: $ echo 1 > /sys/bus/cxl/devices/mem0/trigger_poison_list Example Trace Events: Poison found in a PMEM Region: cxl_poison: memdev=mem0 host=cxl_mem.0 serial=0 trace_type=List region=region11 region_uuid=d96e67ec-76b0-406f-8c35-5b52630dcad1 hpa=0xf100000000 dpa=0x70000000 dpa_length=0x40 source=Injected flags= overflow_time=0 Poison found in RAM Region: cxl_poison: memdev=mem0 host=cxl_mem.0 serial=0 trace_type=List region=region2 region_uuid=00000000-0000-0000-0000-000000000000 hpa=0xf010000000 dpa=0x0 dpa_length=0x40 source=Injected flags= overflow_time=0 Poison found in an unmapped DPA resource: cxl_poison: memdev=mem3 host=cxl_mem.3 serial=3 trace_type=List region= region_uuid=00000000-0000-0000-0000-000000000000 hpa=0xffffffffffffffff dpa=0x40000000 dpa_length=0x40 source=Injected flags= overflow_time=0 [1]: https://www.computeexpresslink.org/download-the-specification Alison Schofield (8): cxl/mbox: Restrict poison cmds to debugfs cxl_raw_allow_all cxl/mbox: Initialize the poison state cxl/mbox: Add GET_POISON_LIST mailbox command cxl/trace: Add TRACE support for CXL media-error records cxl/memdev: Add trigger_poison_list sysfs attribute cxl/region: Provide region info to the cxl_poison trace event cxl/trace: Add an HPA to cxl_poison trace events tools/testing/cxl: Mock support for Get Poison List Dan Williams (1): cxl/mbox: Deprecate poison commands Documentation/ABI/testing/sysfs-bus-cxl | 14 +++ drivers/cxl/core/core.h | 9 ++ drivers/cxl/core/mbox.c | 150 ++++++++++++++++++++++-- drivers/cxl/core/memdev.c | 54 +++++++++ drivers/cxl/core/region.c | 124 ++++++++++++++++++++ drivers/cxl/core/trace.c | 94 +++++++++++++++ drivers/cxl/core/trace.h | 101 ++++++++++++++++ drivers/cxl/cxlmem.h | 83 ++++++++++++- drivers/cxl/mem.c | 43 +++++++ drivers/cxl/pci.c | 4 + include/uapi/linux/cxl_mem.h | 35 +++++- tools/testing/cxl/test/mem.c | 42 +++++++ 12 files changed, 740 insertions(+), 13 deletions(-) base-commit: e686c32590f40bffc45f105c04c836ffad3e531a