<feed xmlns='http://www.w3.org/2005/Atom'>
<title>open-iov.git/tools/guest-diag/run_cycle.sh, branch gr-forwarding</title>
<subtitle>Open-IOV vGPU-IO: host-side GSP emulation for unmodified NVIDIA guest drivers
</subtitle>
<id>https://git.vgpu.io/open-iov.git/atom?h=gr-forwarding</id>
<link rel='self' href='https://git.vgpu.io/open-iov.git/atom?h=gr-forwarding'/>
<link rel='alternate' type='text/html' href='https://git.vgpu.io/open-iov.git/'/>
<updated>2026-10-02T19:13:06+00:00</updated>
<entry>
<title>kernel: GR context-promote bisection modes 4-17 and server-managed-ctxbuf registry variant; complete vgsp rename; docs/tooling refresh</title>
<updated>2026-10-02T19:13:06+00:00</updated>
<author>
<name>Arthur Hanson Rasmusson</name>
<email>arthur@vgpu.io</email>
</author>
<published>2026-10-02T19:13:06+00:00</published>
<link rel='alternate' type='text/html' href='https://git.vgpu.io/open-iov.git/commit/?id=c9bf1cf098c8ea4daeb3e6c9b274a9025c03045d'/>
<id>urn:sha1:c9bf1cf098c8ea4daeb3e6c9b274a9025c03045d</id>
<content type='text'>
- gsp_phys_booter: GR promote bisection modes 4-17 (two-phase zero-field VA,
  batched entries, hVirtMemory, bare VA, KMD SKIP_CTXBUFFER_ALLOC channel,
  old-interface GPU_INITIALIZE_CTX, FBMEM context buffer, UVM-faithful PA-only,
  PATCH/PM auxiliary promotes, server-managed no-promote path); PATCH/PM
  geometry queries and buffers; pre-boot SET_REGISTRY profile selection
- gsp_phys_gpu: gsp_phys_rmcall_gr_promote clamp 0..17 and gsp_phys_srv_ctxbuf
  parameter; registry image variant with RMSetClientRMAllocatedCtxBuffer=0
- vgsp rename across kernel, scripts, tools and docs (module vgsp.ko,
  /sys/kernel/debug/vgsp/, vgsp-&lt;subsystem&gt;: log prefixes)
- risc-v-gsp-rm-debugger user-facing rename and docs/CI refresh; guest-diag
  tool updates; README/SBOM/install/Makefile sync

Live status: CE ladder and P6 guest-copy forwarding green; GR engine context
commit still open (all promote shapes accepted or cleanly refused; the channel
stays CTXNOTVALID) - see Handoff-vGPU-IO.md Addendum 21 (local).
</content>
</entry>
<entry>
<title>guest-diag: stream the client's high-VA UVM sysmem pages as pins</title>
<updated>2026-09-30T16:54:15+00:00</updated>
<author>
<name>Arthur Hanson Rasmusson</name>
<email>arthur@vgpu.io</email>
</author>
<published>2026-09-30T16:54:15+00:00</published>
<link rel='alternate' type='text/html' href='https://git.vgpu.io/open-iov.git/commit/?id=f0d1b77f54a07c249a06dc57280a92d412c01fab'/>
<id>urn:sha1:f0d1b77f54a07c249a06dc57280a92d412c01fab</id>
<content type='text'>
The DtoH staging buffer lives in the client's high-VA /dev/zero|nvidia
mappings (e.g. devptr+0x3c00100), outside the scanner's fixed CE/GR
windows; only the host's pagemap read can see it. The guest scanner
now walks those VMAs and streams their present pages so the module
pins them. run_cycle.sh learns the sema_dbg master switch (auto-on
when any watch knob is set).
</content>
</entry>
<entry>
<title>sema-svc: run CE work before its release, grace unreadable CE entries, payload-pattern audit</title>
<updated>2026-09-30T09:58:38+00:00</updated>
<author>
<name>Arthur Hanson Rasmusson</name>
<email>arthur@vgpu.io</email>
</author>
<published>2026-09-30T09:58:38+00:00</published>
<link rel='alternate' type='text/html' href='https://git.vgpu.io/open-iov.git/commit/?id=06a424b52bf345cd463a2a41d1ac8213df987c02'/>
<id>urn:sha1:06a424b52bf345cd463a2a41d1ac8213df987c02</id>
<content type='text'>
- CE LAUNCH_DMA releases are now applied after the push's collected work
  runs: the client reads copied data as soon as it observes the release,
  so release-first let cuMemcpyDtoH observe a copy that had not executed.
- Unreadable CE entries stay pending for a grace window (sema_ce_grace_ms,
  default 8s) before the release fallback consumes them, so a translation
  that lands within a service pass lets the push be decoded and its work
  executed instead of being dropped.
- sema_copy_pat matches an 8-byte value against every copy's src/dst first
  word and scans decoded pushes for it, naming the submission that carries
  a known payload regardless of size/encoding.
- sema-dma-skip logs transfer-shaped launches that collect no op; op log
  lines carry the channel handle; LAUNCH_DMA inventory follows OPS_LOG.
- A real release application now records the release VA, so decoded pushes
  no longer trigger a spurious fake release.
</content>
</entry>
<entry>
<title>guest-diag: cdt CDT_WAIT park-at-end (live memory inspection); run_cycle CDT_WAIT passthrough</title>
<updated>2026-09-30T09:38:31+00:00</updated>
<author>
<name>Arthur Hanson Rasmusson</name>
<email>arthur@vgpu.io</email>
</author>
<published>2026-09-30T09:38:31+00:00</published>
<link rel='alternate' type='text/html' href='https://git.vgpu.io/open-iov.git/commit/?id=a213fd708dac4494b796bcc9becee1ad4e598b49'/>
<id>urn:sha1:a213fd708dac4494b796bcc9becee1ad4e598b49</id>
<content type='text'>
</content>
</entry>
<entry>
<title>sema-svc: sema_copy_watch (size-keyed src/dst byte audit for the CE data path)</title>
<updated>2026-09-30T09:23:28+00:00</updated>
<author>
<name>Arthur Hanson Rasmusson</name>
<email>arthur@vgpu.io</email>
</author>
<published>2026-09-30T09:23:28+00:00</published>
<link rel='alternate' type='text/html' href='https://git.vgpu.io/open-iov.git/commit/?id=e5d1b3d6ab61b85b945400fcbd3637052feb2ace'/>
<id>urn:sha1:e5d1b3d6ab61b85b945400fcbd3637052feb2ace</id>
<content type='text'>
</content>
</entry>
<entry>
<title>sema-svc: sema_ops_log param (default 8) to audit executed CE operands; run_cycle OPS_LOG knob</title>
<updated>2026-09-30T09:13:03+00:00</updated>
<author>
<name>Arthur Hanson Rasmusson</name>
<email>arthur@vgpu.io</email>
</author>
<published>2026-09-30T09:13:03+00:00</published>
<link rel='alternate' type='text/html' href='https://git.vgpu.io/open-iov.git/commit/?id=50440e5784722b2677eccfbfed391cd880819699'/>
<id>urn:sha1:50440e5784722b2677eccfbfed391cd880819699</id>
<content type='text'>
</content>
</entry>
<entry>
<title>guest-diag: run_cycle enables semfix pushbuffer chain walk by default (CE window has no CPU alias; the vsock stream cannot supply it)</title>
<updated>2026-09-30T08:58:22+00:00</updated>
<author>
<name>Arthur Hanson Rasmusson</name>
<email>arthur@vgpu.io</email>
</author>
<published>2026-09-30T08:58:22+00:00</published>
<link rel='alternate' type='text/html' href='https://git.vgpu.io/open-iov.git/commit/?id=271dda4ac0f8a02d35e97639e2748c51324750e1'/>
<id>urn:sha1:271dda4ac0f8a02d35e97639e2748c51324750e1</id>
<content type='text'>
</content>
</entry>
<entry>
<title>guest-diag: run_cycle SKIP_OPS knob + cdt output on serial; cdt richer roundtrip diagnostics</title>
<updated>2026-09-30T08:41:06+00:00</updated>
<author>
<name>Arthur Hanson Rasmusson</name>
<email>arthur@vgpu.io</email>
</author>
<published>2026-09-30T08:41:06+00:00</published>
<link rel='alternate' type='text/html' href='https://git.vgpu.io/open-iov.git/commit/?id=8bdad7a95d6e6017da4bf6c8ca50e68c4fbcffad'/>
<id>urn:sha1:8bdad7a95d6e6017da4bf6c8ca50e68c4fbcffad</id>
<content type='text'>
</content>
</entry>
<entry>
<title>guest-diag: run_cycle gdb-launch mode + bpz18 early path trace</title>
<updated>2026-09-30T07:24:30+00:00</updated>
<author>
<name>Arthur Hanson Rasmusson</name>
<email>arthur@vgpu.io</email>
</author>
<published>2026-09-30T07:24:30+00:00</published>
<link rel='alternate' type='text/html' href='https://git.vgpu.io/open-iov.git/commit/?id=4604fd5e29ef599aba8f763742a5a8f33be95bb3'/>
<id>urn:sha1:4604fd5e29ef599aba8f763742a5a8f33be95bb3</id>
<content type='text'>
</content>
</entry>
<entry>
<title>guest-diag: run_cycle - verify live gdb before declaring arm failure</title>
<updated>2026-09-30T06:34:16+00:00</updated>
<author>
<name>Arthur Hanson Rasmusson</name>
<email>arthur@vgpu.io</email>
</author>
<published>2026-09-30T06:34:16+00:00</published>
<link rel='alternate' type='text/html' href='https://git.vgpu.io/open-iov.git/commit/?id=a1af81049be12531bf08412edccd83e921414743'/>
<id>urn:sha1:a1af81049be12531bf08412edccd83e921414743</id>
<content type='text'>
</content>
</entry>
</feed>
