diff options
author | Stewart Smith <stewart@linux.ibm.com> | 2018-07-19 19:52:42 +1000 |
---|---|---|
committer | Stewart Smith <stewart@linux.ibm.com> | 2018-07-19 19:53:10 +1000 |
commit | c1d43fc84a30828ac93a20275984dcd6c7608b7a (patch) | |
tree | cec4164a578aa8fc5584ec0bd7c9b412ec65e697 /doc/release-notes | |
parent | 5690c5a8980faf9e528df65dd95535e21c2c868f (diff) | |
download | skiboot-c1d43fc84a30828ac93a20275984dcd6c7608b7a.zip skiboot-c1d43fc84a30828ac93a20275984dcd6c7608b7a.tar.gz skiboot-c1d43fc84a30828ac93a20275984dcd6c7608b7a.tar.bz2 |
skiboot 6.0.6 release notes
Signed-off-by: Stewart Smith <stewart@linux.ibm.com>
(cherry picked from commit 89160502f3695216e9d801b1e97aeee9188a132e)
Signed-off-by: Stewart Smith <stewart@linux.ibm.com>
Diffstat (limited to 'doc/release-notes')
-rw-r--r-- | doc/release-notes/skiboot-6.0.6.rst | 51 |
1 files changed, 51 insertions, 0 deletions
diff --git a/doc/release-notes/skiboot-6.0.6.rst b/doc/release-notes/skiboot-6.0.6.rst new file mode 100644 index 0000000..311586a --- /dev/null +++ b/doc/release-notes/skiboot-6.0.6.rst @@ -0,0 +1,51 @@ +.. _skiboot-6.0.6: + +============= +skiboot-6.0.6 +============= + +skiboot 6.0.6 was released on Thursday July 19th, 2018. It replaces +:ref:`skiboot-6.0.5` as the current stable release in the 6.0.x series. + +It is recommended that 6.0.5 be used instead of any previous 6.0.x version, +especially in the case where NVLINK2 GPUs and/or Mellanox CX5 adapters are +being used. + +Over :ref:`skiboot-6.0.5` we have several important performance related bug +fixes and one stability bug fix: + +- phb4/CAPI: Reallocate PEC2 DMA-Read engines to improve GPU-Direct bandwidth + + We reallocate additional 16/8 DMA-Read engines allocated to stack0/1 + on PEC2 respectively. This is needed to improve bandwidth available to + the Mellanox CX5 adapter when trying to read GPU memory (GPU-Direct). + + If kernel cxl driver indicates a request to allocate maximum possible + DMA read engines when calling enable_capi_mode() and card is attached + to PEC2/stack0 slot then we assume its a Mellanox CX5 adapter. We then + allocate additional 16/8 extra DMA read engines to stack0 and stack1 + respectively on PEC2. This is done by populating the + XPEC_PCI_PRDSTKOVR and XPEC_NEST_READ_STACK_OVERRIDE as suggested by + the h/w team. +- phb4: Disable nodal scoped DMA accesses when PB pump mode is enabled + + By default when a PCIe device issues a read request via the PHB it is first + issued with nodal scope. When accessing GPU memory the NPU does not know at the + time of response if the requested memory page is off node or not. Therefore + every read of GPU memory by a PHB is retried with larger scope which introduces + bandwidth and latency issues. + + On smaller boxes which have pump mode enabled nodal and group scoped reads are + treated the same and both types of request are broadcast to one chip. Therefore + we can avoid the retry by disabling nodal scope on the PHB for these boxes. On + larger boxes nodal (single chip) and group (multiple chip) scoped reads are + treated differently. Therefore we avoid disabling nodal scope on large boxes + which have pump mode disabled to avoid all PHB requests being broadcast to + multiple chips. +- npu2/hw-procedures: Enable parity and credit overflow checks + + Enable these error checking features by setting the appropriate bits in + our one-off initialization of each "NTL Misc Config 2" register. + + The exception is NDL RX parity checking, which should be disabled during + the link training procedures. |