Repository navigation
Conversation
|
The production boot stub pins at macOS 27.2 (26B123), so please retarget this to 27.2. This admits the J615 legacy tree with the 14.x PMP image (OS firmware 14.7); please target the J615 on the 27.2 stub. |
|
Claude Code wrote this up from my J615's firmware versions and the installer's pin list. I'm fine moving this to 27.2, but I can't test it there yet. The floor in this PR was tested on my J615 with the 14.7 stub on system firmware 26.6.2 (mBoot-18000.161.10). The 27.2 stub needs system firmware 27.2 first. That update can't be rolled back. The 26B123 entry in aurora-installer's Is a J615 entry for 26B123 planned? Once there is one, I can update to 27.2, check which PMP image the stub loads and redo the floor against it. Until then, would you rather keep this open for the 14.7 stub or have me close it? |
|
Claude Code wrote this up from my tests. A data point from 7.3 on my J615 (15-inch M3 Air), with this PR, #2090 and #2145 to #2157 on integration/aurora The fault holds until the next boot, since A retry, or re-arming after a minute, would limit a one-off like this to a short gap. I can add that to this PR if you want. |
|
Thanks for getting the memory floor running on the J615, and for the bandwidth numbers. Two parts don't match how we usually select behaviour. The path is limited to one machine by The device tree already tells the cases apart: the legacy tree's report node ( We can run no-regression rows on our M2 if that would help. |
|
j615_floor_eligible() in pmp-report.c keys the PACC0 floor on of_machine_is_compatible("apple,j615") and also needs the new j615_floor module parameter, which defaults to off. So no J615 gets the floor on a default boot. The code already checks the report compatible and the PMP image UUID, and those are what should select the floor, without a board check. Once the floor is qualified it should be on by default with no parameter. Until it is qualified it should stay out of the tree. Until then this should stay a draft. |
|
@freber, a tidy-up pass over the open PRs here. These of yours need something:
If any of these are no longer needed, closing them helps; otherwise a merge or rebase onto integration/aurora is usually all they need. Thanks! |
The 15-inch M3 MacBook Air (J615) has the J613's SoC, DRAM, PMP tables and PACC0 request layout. Its device tree is a legacy handoff tree. The report binds the generic "apple,t8122-pmp-v2-report" and names its PMP with apple,pmp. This driver starts that PMP, which runs the 14.x firmware (3B18C886-4C70-349B-B1C4-70F35DE2C5CD). Without a memory floor nothing raises the DRAM or fabric state for the CPUs. A 4-thread membw2 read then reaches 19.8 GB/s. Give its report apple,cpu-memory-floor = <0x21 2 0x25 4>. The levels come from the J615 runs of the earlier J615 floor patch, not from the J613 tree. That patch stored DCS 4 in bits 51:48 and fabric level 2 in bits 35:32 of the PACC0 request, word 0x2004000200000000. Those are the lanes of request classes 0x25 and 0x21, so these pairs build the same word. The 14.x PMP answered it with DVFS-STATE 0x24 within 5.1 ms. That is class 0x25 at 4 in the low nibble with class 0x21 at 2 in the high nibble. In one boot it confirmed 39 of 39 raises and 38 of 38 releases. With the floor the same membw2 read reached 68.2 GB/s. The levels equal the J613's because the SoC, DRAM and PMP tables are the same, but they were measured on this board. The driver needs no change. pmp_floor_probe() runs for any report with the property. pmp_floor_find_pmp() takes the report's apple,pmp without an availability check, so it does not matter that the PMP is still disabled when the report probes. The start work waits for the report to read READY. It does once the PMP runs, which apple_pmp_start_init() arranges when the display asks for it. Admission then reads the tables m1n1 fills in on the PMP node. Run on this board's live tables, the upstream check admits 2 CPU clients and puts the PACC0 update alias at +0x10888, where the earlier patch wrote it. pmp_floor_word_of() builds 0x2004000200000000 from the pairs. The binding already allows the property on apple,t8122-pmp-v2-report. The generic T8122 report does not set floor_confirm. Raises and releases are published without the DVFS-STATE check, as on every report but the J613 25G83 one. The experimental t8122-j615-25g83 tree includes this file. It binds the native J613 report, where the floor has not run on a J615. It deletes the property so that it stays uncovered, as it was before. Tested by building both J615 trees. Decompiled, the J615 report carries the property while the 25G83 report has none. pmp-report.o builds with clang 18. The floor ran on a J615 with the 14.x firmware on a 7.2.11 kernel carrying the earlier patch. A 7.3 kernel built from this tree with aurora-silicon#2090, aurora-silicon#2199 and aurora-silicon#2203 merged then booted on a J615 with the 14.x firmware and the default command line. The report admitted the floor from the property with the PACC0 update alias at +0x10888. During three 4-thread membw2 reads the floor was up, DVFS-STATE read 0x24 each time and the read reached 69.3 GB/s. The boot logged no floor fault or refusal. On a later boot of the same kernel the read gave 21.6 GB/s with the floor's enable attribute at 0 and 69.3 GB/s with it at 1. Signed-off-by: Fredrik Silfver <1684291+freber@users.noreply.github.com>
4eded98 to
7428f77
Compare
…mplete die key list States 6 to 8 were reachable only with asahi.t8122_pstate_cap and asahi.m3_thermal=on. The measured behaviour needed two more boot parameters. A machine should not need boot options to run as intended. The behaviour belongs to the hardware description. Make the ceiling default to 8 on a T8122 whose device tree has a node with apple,soc-die-keys-complete while the thermal limit is on, which is the default. The property says that the firmware provides every key in apple,soc-die-keys. The hwmon driver then registers the macsmc_soc_die zone only with all of them present. In the tree only the J615 sets it. J613 and J504 list the same keys through t8122.dtsi but are not confirmed to provide them all, so they keep the old ceiling of 2. So does an unset cap with asahi.m3_thermal=off or hold. asahi.t8122_pstate_cap still sets any ceiling from 1 to 8 and is now only an override. The ceiling goes to the firmware in the InitData before the GPU starts, when the hwmon driver may not have probed yet. So the default follows the property, not whether the zone exists at that point. The runtime cap still starts at 2 and falls back to 2. It rises only while the limit reads a usable temperature from that zone, so a J615 whose zone did not register stays at 2. The limit's thresholds follow the same rule. With the property, hot defaults to 90 C and the cool gap to 3 C. Elsewhere the old 80 C and 10 C stay, so the other M3 boards are unchanged. asahi.m3_thermal_hot and asahi.m3_thermal_cool_gap are overrides. Sysfs shows "default" while they are not given. The choice of 90 C with a gap of 3 C comes from the J615 measurements. hot, gap throughput sustained 600 s combined peak 80, 10 218 fps 80.0 fps 95.8 C 85, 10 272 fps 79.9 fps not run 90, 3 289 fps 88.1 fps 94.4 C 92, 3 295 fps 95.2 fps 93.9 C At 90 C with a gap of 3 C a 600 s GPU load held the hottest GPU key at 93.8 C with no missed frames. With a gap of 10 C the cap never left state 2 under a sustained load, because the die settles a few degrees under hot but above hot minus 10. The 85 C of the released form gave no sustained gain while 80 C cost throughput. 92 C is a little faster, but the GPU keys then reach 95.9 C, above the 94 C GPU die target in the boot firmware's tree. 90 C is the highest value that stays under it in the measurements. Critical is then 98 C, below the 110 C trip of the CPU thermal zones. Critical or any loss of a reading returns the cap to 2. This was measured on a J615 only, the one board that sets the property. A 7.3 kernel built from this branch with aurora-silicon#1985, aurora-silicon#2090 and aurora-silicon#2199 merged booted there with the default command line. The cap rose from 2 to 8 once the die reading arrived and gpubench w4 gave 322 fps. Of three 600 s w10 loads at the default, two ran to the end at 126 and 122 fps with the hottest die key at 93.0 C. The third stopped at the test's 95 C guard after 416 s when one die reading reached 99.2 C, where the 98 C step returned the cap to 2. A half-second trace of the same jump on a later load showed a one-second burst on every CPU core, which lifted the CPU keys by 13 to 16 C while the GPU keys rose about 4 C. The host tests in tools/asahi/m3-runtime-tests.rs cover the new defaults and that the old defaults remain without the property. Signed-off-by: Fredrik Silfver <1684291+freber@users.noreply.github.com>
The M3 runtime publishes one render pass at a time unless asahi.m3_render_batch_size or the writable m3_render_batch_override asks for more. On a J615 (15-inch M3 MacBook Air) that costs frames while Chromium scrolls an image-heavy page. Make the boot default 2. Measured on a J615 with the stock 7.2.11 aurora16 module, whose next_batch() and render_batch_size() match this tree's, booted with asahi.t8122_pstate_cap=5 and setting m3_render_batch_override at runtime. Chromium images scene, median of three runs: batch dropped frames p99 frame interval 1 2.2 % 33.3 ms 2 0 % 16.7 ms 4 0.1 % 16.7 ms Two same-session control runs at batch 1 dropped 2.6 % and 2.4 %. With 2, a burst of eight glprobe draws retired 1.6 render commands per retirement instead of 1.0, its median fell from 4.53 ms to 4.00 ms and GPU mailbox interrupts per run fell from 320 to 200. gpubench throughput and idle power stayed within noise. The kernel log had no GPU errors. Only T8122 is affected. This parameter exists only with CONFIG_DRM_ASAHI_M3_T8122, its only reader is the runtime's next_batch(), the runtime's only SoC table is T8122 and it starts only on the J613 and J615. T6030 and T6031 keep the G15 backend or the probe refusal. The first render draw is still published alone for InitBM, and the T8122 pipeline depth of 2 and the 4000 us batch budget still bound every batch. m3_render_batch_override=1 restores the old behaviour at runtime, asahi.m3_render_batch_size=1 at boot. A 7.3 kernel built from this head with aurora-silicon#1985, aurora-silicon#2199 and aurora-silicon#2203 merged then booted on a J615 with the default command line. In the same Chromium scene under the balanced profile at panel scale 1.6, median of three runs, 11.8 % of frames dropped at the default of 2 against 15.6 % on the same kernel booted with asahi.m3_render_batch_size=1. The eight-draw glprobe burst retired 1.52 render commands per retirement against 1.00. The kernel log had no GPU errors. aurora-silicon#2203 raises that build's ceiling to state 8. The J613 runs the same runtime and gets the same default but was not tested. Neither was a 7.3 build at the T8122 ceiling of 2 that applies without aurora-silicon#2203. Signed-off-by: Fredrik Silfver <1684291+freber@users.noreply.github.com>
|
Claude Code wrote this up from my tests. Thanks for #2529. With the floor selected by the device tree, this PR is down to one dts commit on integration/aurora Your comment from 20 November is about I booted it twice on my J615 (15-inch M3 Air). The kernel is integration/aurora
The policy is now your fixed one with the GPU at 40 percent and a 500 ms hold, where my 7.2 tree used 20 percent and 3000 ms. The read with the floor is close to what my 7.2 floor gave (68.2 GB/s). With confirm 0 on this report, the driver publishes raises without checking DVFS-STATE. The fault I posted on 18 November came from that check and can't stop the floor here. To see whether the PMP follows the floor anyway, the test read DVFS-STATE while each run held the floor up. All 6 reads showed 0x24, which is class 0x25 at 4 and class 0x21 at 2. With the check on, one unconfirmed raise cost the floor for the rest of that boot, so I'd keep confirm off for the J615. HW row: J615, integration/aurora The failed unit is speakersafetyd. My boot loader passes no amp levels, so the kernel turns the speakers off. The two new kernel errors were logged during boot, before the tests started. The video encoder finds no preloaded firmware and the camera firmware doesn't answer. Neither touches the PMP or the floor. Thanks for offering M2 rows. Only the J615 dts changes now and the driver stays as #2529 left it, so I don't think they're needed. On the 27.2 retarget, the 26B123 pin in aurora-installer's The title and description now match the single commit. The commit carries only my Signed-off-by and the description ends with |
|
Thanks, this is much simpler now.
Two rows are still missing:
Assisted-by: LLM (Claude:Opus 5.5) |
This gives the J615 (15-inch M3 MacBook Air) PMP report
apple,cpu-memory-floor = <0x21 2 0x25 4>, so the floor from #2529 runs on it by default. The driver doesn't change.The J615 boots m1n1's legacy handoff tree. Its report binds the generic
apple,t8122-pmp-v2-reportand names its PMP withapple,pmp. That PMP runs the 14.x image (3B18C886-4C70-349B-B1C4-70F35DE2C5CD, from OS firmware 14.7). The floor's start work waits for the report to read READY. Admission then reads the tables m1n1 fills in on the PMP node, which on this board gives 2 CPU clients with the PACC0 update alias at +0x10888.The levels are the ones my earlier J615 floor patch used. They build the word 0x2004000200000000, which the PMP answers with DVFS-STATE 0x24. They equal the J613's because the SoC, DRAM and PMP tables are the same, but they were measured on this board.
t8122-j615-25g83.dtsincludes this file and binds the native J613 report, where the floor has never run on a J615. It deletes the property so that tree keeps running without the floor, as before.The generic T8122 report doesn't set floor_confirm, so raises are published without the DVFS-STATE check, as on every report but the J613 25G83 one.
Results on a J615
A 7.3 kernel built from integration/aurora
24dc42277with this commit, #2090, #2199 and #2203 merged, booted twice on the default command line.enablewritten to 0 (3 runs)DVFS-STATE read 0x24 in 6 of 6 reads with the floor up and 0x0 with it off. Neither boot logged a floor fault or refusal.
Not tested
Assisted-by: LLM (Claude:Opus 5.5)