kernel/git/stable/linux.git/Documentation/arm64, branch linux-4.19.y

arm64: errata: Expand speculative SSBS workaround once more

2024-11-08T15:19:15+00:00

[ Upstream commit 081eb7932c2b244f63317a982c5e3990e2c7fbdd ] A number of Arm Ltd CPUs suffer from errata whereby an MSR to the SSBS special-purpose register does not affect subsequent speculative instructions, permitting speculative store bypassing for a window of time. We worked around this for a number of CPUs in commits: * 7187bb7d0b5c7dfa ("arm64: errata: Add workaround for Arm errata 3194386 and 3312417") * 75b3c43eab594bfb ("arm64: errata: Expand speculative SSBS workaround") * 145502cac7ea70b5 ("arm64: errata: Expand speculative SSBS workaround (again)") Since then, a (hopefully final) batch of updates have been published, with two more affected CPUs. For the affected CPUs the existing mitigation is sufficient, as described in their respective Software Developer Errata Notice (SDEN) documents: * Cortex-A715 (MP148) SDEN v15.0, erratum 3456084 https://developer.arm.com/documentation/SDEN-2148827/1500/ * Neoverse-N3 (MP195) SDEN v5.0, erratum 3456111 https://developer.arm.com/documentation/SDEN-3050973/0500/ Enable the existing mitigation by adding the relevant MIDRs to erratum_spec_ssbs_list, and update silicon-errata.rst and the Kconfig text accordingly. Signed-off-by: Mark Rutland Cc: James Morse Cc: Will Deacon Link: https://lore.kernel.org/r/20240930111705.3352047-3-mark.rutland@arm.com Signed-off-by: Catalin Marinas [ Mark: fix conflict in silicon-errata.rst, handle move and rename ] Signed-off-by: Mark Rutland Signed-off-by: Sasha Levin

arm64: errata: Expand speculative SSBS workaround (again)

2024-08-19T03:32:13+00:00

[ Upstream commit adeec61a4723fd3e39da68db4cc4d924e6d7f641 ] A number of Arm Ltd CPUs suffer from errata whereby an MSR to the SSBS special-purpose register does not affect subsequent speculative instructions, permitting speculative store bypassing for a window of time. We worked around this for a number of CPUs in commits: * 7187bb7d0b5c7dfa ("arm64: errata: Add workaround for Arm errata 3194386 and 3312417") * 75b3c43eab594bfb ("arm64: errata: Expand speculative SSBS workaround") Since then, similar errata have been published for a number of other Arm Ltd CPUs, for which the same mitigation is sufficient. This is described in their respective Software Developer Errata Notice (SDEN) documents: * Cortex-A76 (MP052) SDEN v31.0, erratum 3324349 https://developer.arm.com/documentation/SDEN-885749/3100/ * Cortex-A77 (MP074) SDEN v19.0, erratum 3324348 https://developer.arm.com/documentation/SDEN-1152370/1900/ * Cortex-A78 (MP102) SDEN v21.0, erratum 3324344 https://developer.arm.com/documentation/SDEN-1401784/2100/ * Cortex-A78C (MP138) SDEN v16.0, erratum 3324346 https://developer.arm.com/documentation/SDEN-1707916/1600/ * Cortex-A78C (MP154) SDEN v10.0, erratum 3324347 https://developer.arm.com/documentation/SDEN-2004089/1000/ * Cortex-A725 (MP190) SDEN v5.0, erratum 3456106 https://developer.arm.com/documentation/SDEN-2832921/0500/ * Cortex-X1 (MP077) SDEN v21.0, erratum 3324344 https://developer.arm.com/documentation/SDEN-1401782/2100/ * Cortex-X1C (MP136) SDEN v16.0, erratum 3324346 https://developer.arm.com/documentation/SDEN-1707914/1600/ * Neoverse-N1 (MP050) SDEN v32.0, erratum 3324349 https://developer.arm.com/documentation/SDEN-885747/3200/ * Neoverse-V1 (MP076) SDEN v19.0, erratum 3324341 https://developer.arm.com/documentation/SDEN-1401781/1900/ Note that due to the manner in which Arm develops IP and tracks errata, some CPUs share a common erratum number and some CPUs have multiple erratum numbers for the same HW issue. On parts without SB, it is necessary to use ISB for the workaround. The spec_bar() macro used in the mitigation will expand to a "DSB SY; ISB" sequence in this case, which is sufficient on all affected parts. Enable the existing mitigation by adding the relevant MIDRs to erratum_spec_ssbs_list. The list is sorted alphanumerically (involving moving Neoverse-V3 after Neoverse-V2) so that this is easy to audit and potentially extend again in future. The Kconfig text is also updated to clarify the set of affected parts and the mitigation. Signed-off-by: Mark Rutland Cc: James Morse Cc: Will Deacon Reviewed-by: Anshuman Khandual Acked-by: Will Deacon Link: https://lore.kernel.org/r/20240801101803.1982459-4-mark.rutland@arm.com Signed-off-by: Catalin Marinas [ Mark: fix conflicts in silicon-errata.rst ] Signed-off-by: Mark Rutland Signed-off-by: Sasha Levin

arm64: errata: Expand speculative SSBS workaround

2024-08-19T03:32:13+00:00

[ Upstream commit 75b3c43eab594bfbd8184ec8ee1a6b820950819a ] A number of Arm Ltd CPUs suffer from errata whereby an MSR to the SSBS special-purpose register does not affect subsequent speculative instructions, permitting speculative store bypassing for a window of time. We worked around this for Cortex-X4 and Neoverse-V3, in commit: 7187bb7d0b5c7dfa ("arm64: errata: Add workaround for Arm errata 3194386 and 3312417") ... as per their Software Developer Errata Notice (SDEN) documents: * Cortex-X4 SDEN v8.0, erratum 3194386: https://developer.arm.com/documentation/SDEN-2432808/0800/ * Neoverse-V3 SDEN v6.0, erratum 3312417: https://developer.arm.com/documentation/SDEN-2891958/0600/ Since then, similar errata have been published for a number of other Arm Ltd CPUs, for which the mitigation is the same. This is described in their respective SDEN documents: * Cortex-A710 SDEN v19.0, errataum 3324338 https://developer.arm.com/documentation/SDEN-1775101/1900/?lang=en * Cortex-A720 SDEN v11.0, erratum 3456091 https://developer.arm.com/documentation/SDEN-2439421/1100/?lang=en * Cortex-X2 SDEN v19.0, erratum 3324338 https://developer.arm.com/documentation/SDEN-1775100/1900/?lang=en * Cortex-X3 SDEN v14.0, erratum 3324335 https://developer.arm.com/documentation/SDEN-2055130/1400/?lang=en * Cortex-X925 SDEN v8.0, erratum 3324334 https://developer.arm.com/documentation/109108/800/?lang=en * Neoverse-N2 SDEN v17.0, erratum 3324339 https://developer.arm.com/documentation/SDEN-1982442/1700/?lang=en * Neoverse-V2 SDEN v9.0, erratum 3324336 https://developer.arm.com/documentation/SDEN-2332927/900/?lang=en Note that due to shared design lineage, some CPUs share the same erratum number. Add these to the existing mitigation under CONFIG_ARM64_ERRATUM_3194386. As listing all of the erratum IDs in the runtime description would be unwieldy, this is reduced to: "SSBS not fully self-synchronizing" ... matching the description of the errata in all of the SDENs. Signed-off-by: Mark Rutland Cc: James Morse Cc: Will Deacon Link: https://lore.kernel.org/r/20240603111812.1514101-6-mark.rutland@arm.com Signed-off-by: Catalin Marinas [ Mark: fix conflicts and renames ] Signed-off-by: Mark Rutland Signed-off-by: Sasha Levin

arm64: errata: Unify speculative SSBS errata logic

2024-08-19T03:32:13+00:00

[ Upstream commit ec768766608092087dfb5c1fc45a16a6f524dee2 ] Cortex-X4 erratum 3194386 and Neoverse-V3 erratum 3312417 are identical, with duplicate Kconfig text and some unsightly ifdeffery. While we try to share code behind CONFIG_ARM64_WORKAROUND_SPECULATIVE_SSBS, having separate options results in a fair amount of boilerplate code, and this will only get worse as we expand the set of affected CPUs. To reduce this boilerplate, unify the two behind a common Kconfig option. This removes the duplicate text and Kconfig logic, and removes the need for the intermediate ARM64_WORKAROUND_SPECULATIVE_SSBS option. The set of affected CPUs is described as a list so that this can easily be extended. I've used ARM64_ERRATUM_3194386 (matching the Neoverse-V3 erratum ID) as the common option, matching the way we use ARM64_ERRATUM_1319367 to cover Cortex-A57 erratum 1319537 and Cortex-A72 erratum 1319367. Signed-off-by: Mark Rutland Cc: James Morse Cc: Will Deacon Link: https://lore.kernel.org/r/20240603111812.1514101-5-mark.rutland@arm.com Signed-off-by: Catalin Marinas [ Mark: fix conflicts & renames, drop unneeded cpucaps.h ] Signed-off-by: Mark Rutland Signed-off-by: Sasha Levin

arm64: errata: Add workaround for Arm errata 3194386 and 3312417

2024-08-19T03:32:12+00:00

[ Upstream commit 7187bb7d0b5c7dfa18ca82e9e5c75e13861b1d88 ] Cortex-X4 and Neoverse-V3 suffer from errata whereby an MSR to the SSBS special-purpose register does not affect subsequent speculative instructions, permitting speculative store bypassing for a window of time. This is described in their Software Developer Errata Notice (SDEN) documents: * Cortex-X4 SDEN v8.0, erratum 3194386: https://developer.arm.com/documentation/SDEN-2432808/0800/ * Neoverse-V3 SDEN v6.0, erratum 3312417: https://developer.arm.com/documentation/SDEN-2891958/0600/ To workaround these errata, it is necessary to place a speculation barrier (SB) after MSR to the SSBS special-purpose register. This patch adds the requisite SB after writes to SSBS within the kernel, and hides the presence of SSBS from EL0 such that userspace software which cares about SSBS will manipulate this via prctl(PR_GET_SPECULATION_CTRL, ...). Signed-off-by: Mark Rutland Cc: Catalin Marinas Cc: James Morse Cc: Will Deacon Link: https://lore.kernel.org/r/20240508081400.235362-5-mark.rutland@arm.com Signed-off-by: Will Deacon [ Mark: fix conflicts & renames, drop unneeded cpucaps.h, fold in user_feature_fixup() ] Signed-off-by: Mark Rutland Signed-off-by: Sasha Levin

arm64: errata: Remove AES hwcap for COMPAT tasks

2022-11-03T14:52:25+00:00

commit 44b3834b2eed595af07021b1c64e6f9bc396398b upstream. Cortex-A57 and Cortex-A72 have an erratum where an interrupt that occurs between a pair of AES instructions in aarch32 mode may corrupt the ELR. The task will subsequently produce the wrong AES result. The AES instructions are part of the cryptographic extensions, which are optional. User-space software will detect the support for these instructions from the hwcaps. If the platform doesn't support these instructions a software implementation should be used. Remove the hwcap bits on affected parts to indicate user-space should not use the AES instructions. Acked-by: Ard Biesheuvel Signed-off-by: James Morse Link: https://lore.kernel.org/r/20220714161523.279570-3-james.morse@arm.com Signed-off-by: Will Deacon [florian: resolved conflicts in arch/arm64/tools/cpucaps and cpu_errata.c] Signed-off-by: Florian Fainelli Signed-off-by: Greg Kroah-Hartman

arm64: errata: Hide CTR_EL0.DIC on systems affected by Neoverse-N1 #1542419

2020-04-29T14:31:08+00:00

[ Upstream commit 05460849c3b51180d5ada3373d0449aea19075e4 ] Cores affected by Neoverse-N1 #1542419 could execute a stale instruction when a branch is updated to point to freshly generated instructions. To workaround this issue we need user-space to issue unnecessary icache maintenance that we can trap. Start by hiding CTR_EL0.DIC. Reviewed-by: Suzuki K Poulose Signed-off-by: James Morse Signed-off-by: Catalin Marinas [ Removed cpu_enable_trap_ctr_access() hunk due to no 4afe8e79da92] Signed-off-by: James Morse Signed-off-by: Sasha Levin

arm64: docs: Document SSBS HWCAP

2019-10-11T16:21:31+00:00

[ Upstream commit ee91176120bd584aa10c564e7e9fdcaf397190a1 ] We advertise the MRS/MSR instructions for toggling SSBS at EL0 using an HWCAP, so document it along with the others. Signed-off-by: Will Deacon Signed-off-by: Catalin Marinas Signed-off-by: Ard Biesheuvel Signed-off-by: Greg Kroah-Hartman

arm64: errata: Add workaround for Cortex-A76 erratum #1463225

2019-05-31T13:46:03+00:00

commit 969f5ea627570e91c9d54403287ee3ed657f58fe upstream. Revisions of the Cortex-A76 CPU prior to r4p0 are affected by an erratum that can prevent interrupts from being taken when single-stepping. This patch implements a software workaround to prevent userspace from effectively being able to disable interrupts. Cc: Cc: Marc Zyngier Cc: Catalin Marinas Signed-off-by: Will Deacon Signed-off-by: Greg Kroah-Hartman

clocksource/drivers/arch_timer: Workaround for Allwinner A64 timer instability

2019-03-23T19:09:58+00:00

commit c950ca8c35eeb32224a63adc47e12f9e226da241 upstream. The Allwinner A64 SoC is known[1] to have an unstable architectural timer, which manifests itself most obviously in the time jumping forward a multiple of 95 years[2][3]. This coincides with 2^56 cycles at a timer frequency of 24 MHz, implying that the time went slightly backward (and this was interpreted by the kernel as it jumping forward and wrapping around past the epoch). Investigation revealed instability in the low bits of CNTVCT at the point a high bit rolls over. This leads to power-of-two cycle forward and backward jumps. (Testing shows that forward jumps are about twice as likely as backward jumps.) Since the counter value returns to normal after an indeterminate read, each "jump" really consists of both a forward and backward jump from the software perspective. Unless the kernel is trapping CNTVCT reads, a userspace program is able to read the register in a loop faster than it changes. A test program running on all 4 CPU cores that reported jumps larger than 100 ms was run for 13.6 hours and reported the following: Count | Event -------+--------------------------- 9940 | jumped backward 699ms 268 | jumped backward 1398ms 1 | jumped backward 2097ms 16020 | jumped forward 175ms 6443 | jumped forward 699ms 2976 | jumped forward 1398ms 9 | jumped forward 356516ms 9 | jumped forward 357215ms 4 | jumped forward 714430ms 1 | jumped forward 3578440ms This works out to a jump larger than 100 ms about every 5.5 seconds on each CPU core. The largest jump (almost an hour!) was the following sequence of reads: 0x0000007fffffffff → 0x00000093feffffff → 0x0000008000000000 Note that the middle bits don't necessarily all read as all zeroes or all ones during the anomalous behavior; however the low 10 bits checked by the function in this patch have never been observed with any other value. Also note that smaller jumps are much more common, with backward jumps of 2048 (2^11) cycles observed over 400 times per second on each core. (Of course, this is partially explained by lower bits rolling over more frequently.) Any one of these could have caused the 95 year time skip. Similar anomalies were observed while reading CNTPCT (after patching the kernel to allow reads from userspace). However, the CNTPCT jumps are much less frequent, and only small jumps were observed. The same program as before (except now reading CNTPCT) observed after 72 hours: Count | Event -------+--------------------------- 17 | jumped backward 699ms 52 | jumped forward 175ms 2831 | jumped forward 699ms 5 | jumped forward 1398ms Further investigation showed that the instability in CNTPCT/CNTVCT also affected the respective timer's TVAL register. The following values were observed immediately after writing CNVT_TVAL to 0x10000000: CNTVCT | CNTV_TVAL | CNTV_CVAL | CNTV_TVAL Error --------------------+------------+--------------------+----------------- 0x000000d4a2d8bfff | 0x10003fff | 0x000000d4b2d8bfff | +0x00004000 0x000000d4a2d94000 | 0x0fffffff | 0x000000d4b2d97fff | -0x00004000 0x000000d4a2d97fff | 0x10003fff | 0x000000d4b2d97fff | +0x00004000 0x000000d4a2d9c000 | 0x0fffffff | 0x000000d4b2d9ffff | -0x00004000 The pattern of errors in CNTV_TVAL seemed to depend on exactly which value was written to it. For example, after writing 0x10101010: CNTVCT | CNTV_TVAL | CNTV_CVAL | CNTV_TVAL Error --------------------+------------+--------------------+----------------- 0x000001ac3effffff | 0x1110100f | 0x000001ac4f10100f | +0x1000000 0x000001ac40000000 | 0x1010100f | 0x000001ac5110100f | -0x1000000 0x000001ac58ffffff | 0x1110100f | 0x000001ac6910100f | +0x1000000 0x000001ac66000000 | 0x1010100f | 0x000001ac7710100f | -0x1000000 0x000001ac6affffff | 0x1110100f | 0x000001ac7b10100f | +0x1000000 0x000001ac6e000000 | 0x1010100f | 0x000001ac7f10100f | -0x1000000 I was also twice able to reproduce the issue covered by Allwinner's workaround[4], that writing to TVAL sometimes fails, and both CVAL and TVAL are left with entirely bogus values. One was the following values: CNTVCT | CNTV_TVAL | CNTV_CVAL --------------------+------------+-------------------------------------- 0x000000d4a2d6014c | 0x8fbd5721 | 0x000000d132935fff (615s in the past) Reviewed-by: Marc Zyngier ======================================================================== Because the CPU can read the CNTPCT/CNTVCT registers faster than they change, performing two reads of the register and comparing the high bits (like other workarounds) is not a workable solution. And because the timer can jump both forward and backward, no pair of reads can distinguish a good value from a bad one. The only way to guarantee a good value from consecutive reads would be to read _three_ times, and take the middle value only if the three values are 1) each unique and 2) increasing. This takes at minimum 3 counter cycles (125 ns), or more if an anomaly is detected. However, since there is a distinct pattern to the bad values, we can optimize the common case (1022/1024 of the time) to a single read by simply ignoring values that match the error pattern. This still takes no more than 3 cycles in the worst case, and requires much less code. As an additional safety check, we still limit the loop iteration to the number of max-frequency (1.2 GHz) CPU cycles in three 24 MHz counter periods. For the TVAL registers, the simple solution is to not use them. Instead, read or write the CVAL and calculate the TVAL value in software. Although the manufacturer is aware of at least part of the erratum[4], there is no official name for it. For now, use the kernel-internal name "UNKNOWN1". [1]: https://github.com/armbian/build/commit/a08cd6fe7ae9 [2]: https://forum.armbian.com/topic/3458-a64-datetime-clock-issue/ [3]: https://irclog.whitequark.org/linux-sunxi/2018-01-26 [4]: https://github.com/Allwinner-Homlet/H6-BSP4.9-linux/blob/master/drivers/clocksource/arm_arch_timer.c#L272 Acked-by: Maxime Ripard Tested-by: Andre Przywara Signed-off-by: Samuel Holland Cc: stable@vger.kernel.org Signed-off-by: Daniel Lezcano Signed-off-by: Greg Kroah-Hartman