- 23 Mar, 2018 9 commits
-
-
Michael Ellerman authored
This brings in two series from Paul, one of which touches KVM code and may need to be merged into the kvm-ppc tree to resolve conflicts.
-
Paul Mackerras authored
This works around a hardware bug in "Nimbus" POWER9 DD2.2 processors, where the contents of the TEXASR can get corrupted while a thread is in fake suspend state. The workaround is for the instruction emulation code to use the value saved at the most recent guest exit in real suspend mode. We achieve this by simply not saving the TEXASR into the vcpu struct on an exit in fake suspend state. We also have to take care to set the orig_texasr field only on guest exit in real suspend state. This also means that on guest entry in fake suspend state, TEXASR will be restored to the value it had on the last exit in real suspend state, effectively counteracting any hardware-caused corruption. This works because TEXASR may not be written in suspend state. With this, the guest might see the wrong values in TEXASR if it reads it while in suspend state, but will see the correct value in non-transactional state (e.g. after a treclaim), and treclaim will work correctly. With this workaround, the code will actually run slightly faster, and will operate correctly on systems without the TEXASR bug (since TEXASR may not be written in suspend state, and is only changed by failure recording, which will have already been done before we get into fake suspend state). Therefore these changes are not made subject to a CPU feature bit. Signed-off-by:
Paul Mackerras <paulus@ozlabs.org> Signed-off-by:
Michael Ellerman <mpe@ellerman.id.au>
-
Suraj Jitindar Singh authored
This works around a hardware bug in "Nimbus" POWER9 DD2.2 processors, where a treclaim performed in fake suspend mode can cause subsequent reads from the XER register to return inconsistent values for the SO (summary overflow) bit. The inconsistent SO bit state can potentially be observed on any thread in the core. We have to do the treclaim because that is the only way to get the thread out of suspend state (fake or real) and into non-transactional state. The workaround for the bug is to force the core into SMT4 mode before doing the treclaim. This patch adds the code to do that, conditional on the CPU_FTR_P9_TM_XER_SO_BUG feature bit. Signed-off-by:
Suraj Jitindar Singh <sjitindarsingh@gmail.com> Signed-off-by:
-
Paul Mackerras authored
POWER9 has hardware bugs relating to transactional memory and thread reconfiguration (changes to hardware SMT mode). Specifically, the core does not have enough storage to store a complete checkpoint of all the architected state for all four threads. The DD2.2 version of POWER9 includes hardware modifications designed to allow hypervisor software to implement workarounds for these problems. This patch implements those workarounds in KVM code so that KVM guests see a full, working transactional memory implementation. The problems center around the use of TM suspended state, where the CPU has a checkpointed state but execution is not transactional. The workaround is to implement a "fake suspend" state, which looks to the guest like suspended state but the CPU does not store a checkpoint. In this state, any instruction that would cause a transition to transactional state (rfid, rfebb, mtmsrd, tresume) or would use the checkpointed state (treclaim) causes a "soft patch" interrupt (vector 0x1500) to the hypervisor so that it can be emulated. The trechkpt instruction also causes a soft patch interrupt. On POWER9 DD2.2, we avoid returning to the guest in any state which would require a checkpoint to be present. The trechkpt in the guest entry path which would normally create that checkpoint is replaced by either a transition to fake suspend state, if the guest is in suspend state, or a rollback to the pre-transactional state if the guest is in transactional state. Fake suspend state is indicated by a flag in the PACA plus a new bit in the PSSCR. The new PSSCR bit is write-only and reads back as 0. On exit from the guest, if the guest is in fake suspend state, we still do the treclaim instruction as we would in real suspend state, in order to get into non-transactional state, but we do not save the resulting register state since there was no checkpoint. Emulation of the instructions that cause a softpatch interrupt is handled in two paths. If the guest is in real suspend mode, we call kvmhv_p9_tm_emulation_early() to handle the cases where the guest is transitioning to transactional state. This is called before we do the treclaim in the guest exit path; because we haven't done treclaim, we can get back to the guest with the transaction still active. If the instruction is a case that kvmhv_p9_tm_emulation_early() doesn't handle, or if the guest is in fake suspend state, then we proceed to do the complete guest exit path and subsequently call kvmhv_p9_tm_emulation() in host context with the MMU on. This handles all the cases including the cases that generate program interrupts (illegal instruction or TM Bad Thing) and facility unavailable interrupts. The emulation is reasonably straightforward and is mostly concerned with checking for exception conditions and updating the state of registers such as MSR and CR0. The treclaim emulation takes care to ensure that the TEXASR register gets updated as if it were the guest treclaim instruction that had done failure recording, not the treclaim done in hypervisor state in the guest exit path. With this, the KVM_CAP_PPC_HTM capability returns true (1) even if transactional memory is not available to host userspace. Signed-off-by:
-
Paul Mackerras authored
POWER9 processors up to and including "Nimbus" v2.2 have hardware bugs relating to transactional memory and thread reconfiguration. One of these bugs has a workaround which is to get the core into SMT4 state temporarily. This workaround is only needed when running bare-metal. This patch provides a function which gets the core into SMT4 mode by preventing threads from going to a stop state, and waking up those which are already in a stop state. Once at least 3 threads are not in a stop state, the core will be in SMT4 and we can continue. To do this, we add a "dont_stop" flag to the paca to tell the thread not to go into a stop state. If this flag is set, power9_idle_stop() just returns immediately with a return value of 0. The pnv_power9_force_smt4_catch() function does the following: 1. Set the dont_stop flag for each thread in the core, except ourselves (in fact we use an atomic_inc() in case more than one thread is calling this function concurrently). 2. See how many threads are awake, indicated by their requested_psscr field in the paca being 0. If this is at least 3, skip to step 5. 3. Send a doorbell interrupt to each thread that was seen as being in a stop state in step 2. 4. Until at least 3 threads are awake, scan the threads to which we sent a doorbell interrupt and check if they are awake now. This relies on the following properties: - Once dont_stop is non-zero, requested_psccr can't go from zero to non-zero, except transiently (and without the thread doing stop). - requested_psscr being zero guarantees that the thread isn't in a state-losing stop state where thread reconfiguration could occur. - Doing stop with a PSSCR value of 0 won't be a state-losing stop and thus won't allow thread reconfiguration. - Once threads_per_core/2 + 1 (i.e. 3) threads are awake, the core must be in SMT4 mode, since SMT modes are powers of 2. This does add a sync to power9_idle_stop(), which is necessary to provide the correct ordering between setting requested_psscr and checking dont_stop. The overhead of the sync should be unnoticeable compared to the latency of going into and out of a stop state. Because some objected to incurring this extra latency on systems where the XER[SO] bug is not relevant, I have put the test in power9_idle_stop inside a feature section. This means that pnv_power9_force_smt4_catch() WILL NOT WORK correctly on systems without the CPU_FTR_P9_TM_XER_SO_BUG feature bit set, and will probably hang the system. In order to cater for uses where the caller has an operation that has to be done while the core is in SMT4, the core continues to be kept in SMT4 after pnv_power9_force_smt4_catch() function returns, until the pnv_power9_force_smt4_release() function is called. It undoes the effect of step 1 above and allows the other threads to go into a stop state. Signed-off-by:
-
Paul Mackerras authored
This adds a CPU feature bit which is set for POWER9 "Nimbus" DD2.2 processors which will be used to enable the hypervisor to assist hardware with the handling of checkpointed register values while the CPU is in suspend state, in order to work around hardware bugs. The hardware assistance for these workarounds introduced a new hardware bug relating to the XER[SO] bit. We add a separate feature bit for this bug in case future chips fix it while still requiring the hypervisor assistance with suspend state. When the dt_cpu_ftrs subsystem is in use, the software assistance can be enabled using a "tm-suspend-hypervisor-assist" node in the device tree, and a "tm-suspend-xer-so-bug" node enables the workarounds for the XER[SO] bug. In the absence of such nodes, a quirk enables both for POWER9 "Nimbus" DD2.2 processors. Signed-off-by:
-
Paul Mackerras authored
This moves all the CPU feature bits that are only used on 32-bit machines to the top 20 bits of the CPU feature word and arranges for them to be defined only in 32-bit builds. The features that are common to 32-bit and 64-bit machines are moved to bits 0-11 of the CPU feature word. This means that for 64-bit platforms, bits 44-63 can now be used for new features that only exist on 64-bit machines. (These bit numbers are counting from the right, i.e. the LSB is bit 0.) Because CPU_FTR_L3_DISABLE_NAP moved from the low 16 bits to the high 16 bits, we have to adjust some assembly code. Also, CPU_FTR_EMB_HV moved from the high 16 bits to the low 16 bits. Note that CPU_FTR_REAL_LE only applies to 64-bit chips, because only 64-bit chips (POWER6, 7, 8, 9) have a true little-endian mode that is a CPU execution mode as opposed to being a page attribute. With this we now have 20 free CPU feature bits on 64-bit machines. Signed-off-by:
-
Paul Mackerras authored
The CPU_FTR_L2CSR bit is never tested anywhere, so let's reclaim the bit. The last usage was removed in 86d63363 ("powerpc/e500mc: Remove dead L2 flushing code in idle_e500.S") (Jun 2015). Signed-off-by:
-
Paul Mackerras authored
All PowerPC CPUs other than the original PPC601 have a timebase register rather than the "real-time clock" (RTC) register that the PPC601 (and the original POWER and POWER2 CPUs) had. Currently we have a CPU feature bit to indicate the presence of the timebase, but it makes more sense to use a bit to indicate the unusual situation rather than the common situation. This therefore defines a CPU_FTR_USE_RTC bit in place of the CPU_FTR_USE_TB bit, and arranges for it to be set on PPC601 systems. Signed-off-by:
-
- 20 Mar, 2018 6 commits
-
-
Rob Herring authored
'linux,stdout-path' has been deprecated for some time in favor of 'stdout-path'. Now dtc will warn on occurrences of 'linux,stdout-path'. Search and replace all the of occurrences with 'stdout-path'. Signed-off-by:
Rob Herring <robh@kernel.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: linuxppc-dev@lists.ozlabs.org Signed-off-by:
-
Nicholas Piggin authored
Signed-off-by:
Nicholas Piggin <npiggin@gmail.com> [mpe: Add SPDX, and fixup formatting] Signed-off-by:
-
Markus Elfring authored
It's slightly less error prone to use sizeof(*foo) rather than specifying the type. Signed-off-by:
Markus Elfring <elfring@users.sourceforge.net> [mpe: Consolidate into one patch, rewrite change log] Signed-off-by:
-
Matt Brown authored
The flush_dcache_phys_range() function is no longer used in the kernel. The last usage was removed in c40785ad ("powerpc/dart: Use a cachable DART"). This patch removes the function and declaration. Signed-off-by:
Matt Brown <matthew.brown.dev@gmail.com> [mpe: Munge change log, include commit that removed last user] Signed-off-by:
-
Matt Brown authored
Previously the raid6 test Makefile did not build the POWER specific files (altivec and vpermxor). This patch fixes the bug, so that all appropriate files for powerpc are built. This patch also fixes the missing and mismatched ifdef statements to allow the altivec.uc file to be built correctly. Signed-off-by:
-
Matt Brown authored
This patch uses the vpermxor instruction to optimise the raid6 Q syndrome. This instruction was made available with POWER8, ISA version 2.07. It allows for both vperm and vxor instructions to be done in a single instruction. This has been tested for correctness on a ppc64le vm with a basic RAID6 setup containing 5 drives. The performance benchmarks are from the raid6test in the /lib/raid6/test directory. These results are from an IBM Firestone machine with ppc64le architecture. The benchmark results show a 35% speed increase over the best existing algorithm for powerpc (altivec). The raid6test has also been run on a big-endian ppc64 vm to ensure it also works for big-endian architectures. Performance benchmarks: raid6: altivecx4 gen() 18773 MB/s raid6: altivecx8 gen() 19438 MB/s raid6: vpermxor4 gen() 25112 MB/s raid6: vpermxor8 gen() 26279 MB/s Signed-off-by:
Daniel Axtens <dja@axtens.net> [mpe: Add VPERMXOR macro so we can build with old binutils] Signed-off-by:
-
- 14 Mar, 2018 6 commits
-
-
Alexandre Belloni authored
The proper compatible for rv3029 is microcrystal,rv3029. Acked-by:
Anatolij Gustschin <agust@denx.de> Signed-off-by:
Alexandre Belloni <alexandre.belloni@bootlin.com> Signed-off-by:
-
Alexandre Belloni authored
The RTC core is always calling rtc_valid_tm after the read_time callback. It is not necessary to call it just before returning from the callback. Signed-off-by:
-
Sukadev Bhattiprolu authored
Add a couple of trace points in the VAS driver Signed-off-by:
Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com> [mpe: Add SPDX tag to new header] Signed-off-by:
-
Sukadev Bhattiprolu authored
When VAS is not configured, unregister the platform driver. Also simplify cleanup by delaying vas debugfs init until we know VAS is configured. Signed-off-by:
-
Mark Hairgrove authored
pnv_npu2_init_context wasn't checking the return code from __mmu_notifier_register. If __mmu_notifier_register failed, the npu_context was still assigned to the mm and the caller wasn't given any indication that things went wrong. Later on pnv_npu2_destroy_context would be called, which in turn called mmu_notifier_unregister and dropped mm->mm_count without having incremented it in the first place. This led to various forms of corruption like mm use-after-free and mm double-free. __mmu_notifier_register can fail with EINTR if a signal is pending, so this case can be frequent. This patch calls opal_npu_destroy_context on the failure paths, and makes sure not to assign mm->context.npu_context until past the failure points. Signed-off-by:
Mark Hairgrove <mhairgrove@nvidia.com> Acked-By:
Alistair Popple <alistair@popple.id.au> Signed-off-by:
-
Christophe Lombard authored
The PSL Timebase register is updated by the PSL to maintain the timebase. On P9, the Timebase value is only provided by the CAPP as received the last time a timebase request was performed. The timebase requests are initiated through the adapter configuration or application registers. The specific sysfs entry "/sys/class/cxl/cardxx/psl_timebase_synced" is now dynamically updated according the content of the PSL Timebase register. Fixes: f24be42a ("cxl: Add psl9 specific code") Signed-off-by:
Christophe Lombard <clombard@linux.vnet.ibm.com> Reviewed-by:
Vaibhav Jain <vaibhav@linux.vnet.ibm.com> Acked-by:
Andrew Donnellan <andrew.donnellan@au1.ibm.com> Acked-by:
Frederic Barrat <fbarrat@linux.vnet.ibm.com> Signed-off-by:
-
- 13 Mar, 2018 19 commits
-
-
Nicholas Piggin authored
This is a tidy up which removes radix MMU calls into the slice code. Signed-off-by:
-
Nicholas Piggin authored
The slice_mask cache was a basic conversion which copied the slice mask into caller's structures, because that's how the original code worked. In most cases the pointer can be used directly instead, saving a copy and an on-stack structure. On POWER8, this increases vfork+exec+exit performance by 0.3% and reduces time to mmap+munmap a 64kB page by 2%. Signed-off-by:
-
Nicholas Piggin authored
This code is never compiled in, and it gets broken by the next patch, so remove it. Signed-off-by:
-
Nicholas Piggin authored
This converts the slice_mask bit operation helpers to be the usual 3-operand kind, which allows 2 inputs to set a different output without an extra copy, which is used in the next patch. Adds slice_copy_mask, which will be used in the next patch. Signed-off-by:
-
Nicholas Piggin authored
Rather than build slice masks from a range then use that to check for fit in a candidate mask, implement slice_check_range_fits that checks if a range fits in a mask directly. This allows several structures to be removed from stacks, and also we don't expect a huge range in a lot of these cases, so building and comparing a full mask is going to be more expensive than testing just one or two bits of the range. On POWER8, this increases vfork+exec+exit performance by 0.3% and reduces time to mmap+munmap a 64kB page by 5%. Signed-off-by:
-
Nicholas Piggin authored
Calculating the slice mask can become a signifcant overhead for get_unmapped_area. This patch adds a struct slice_mask for each page size in the mm_context, and keeps these in synch with the slices psize arrays and slb_addr_limit. On Book3S/64 this adds 288 bytes to the mm_context_t for the slice mask caches. On POWER8, this increases vfork+exec+exit performance by 9.9% and reduces time to mmap+munmap a 64kB page by 28%. Reduces time to mmap+munmap by about 10% on 8xx. Signed-off-by:
-
Nicholas Piggin authored
Pass around const pointers to struct slice_mask where possible, rather than copies of slice_mask, to reduce stack and call overhead. checkstack.pl gives, before: 0x00000d1c slice_get_unmapped_area [slice.o]: 592 0x00001864 is_hugepage_only_range [slice.o]: 448 0x00000754 slice_find_area_topdown [slice.o]: 400 0x00000484 slice_find_area_bottomup.isra.1 [slice.o]: 272 0x000017b4 slice_set_range_psize [slice.o]: 224 0x00000a4c slice_find_area [slice.o]: 128 0x00000160 slice_check_fit [slice.o]: 112 after: 0x00000ad0 slice_get_unmapped_area [slice.o]: 448 0x00001464 is_hugepage_only_range [slice.o]: 288 0x000006c0 slice_find_area [slice.o]: 144 0x0000016c slice_check_fit [slice.o]: 128 0x00000528 slice_find_area_bottomup.isra.2 [slice.o]: 128 0x000013e4 slice_set_range_psize [slice.o]: 128 This increases vfork+exec+exit performance by 1.5%. Reduces time to mmap+munmap a 64kB page by 17%. Signed-off-by:
-
Nicholas Piggin authored
Make these loops look the same, and change their form so the important part is not wrapped over so many lines. Signed-off-by:
-
Nicholas Piggin authored
The slice state of an mm gets zeroed then initialised upon exec. This is the only caller of slice_set_user_psize now, so that can be removed and instead implement a faster and simplified approach that requires no locking or checking existing state. This speeds up vfork+exec+exit performance on POWER8 by 3%. Signed-off-by:
-
Michael Ellerman authored
Now that plpar_wrappers.h has an #ifdef PSERIES we can move the empty version of plpar_set_ciabr() which xmon wants into there. Signed-off-by: -
Michael Ellerman authored
Back in 2013 we added some hypercall wrappers which misspelled "plpar" (P-series Logical PARtition) as "plapr". Visually they're hard to distinguish and it almost doesn't matter, but it is confusing when grepping to miss some calls because of the typo. They've also started spreading, so before they take over let's fix them all to be "plpar". Signed-off-by: -
Michael Ellerman authored
Currently plpar_wrappers.h is not safe to include when CONFIG_PPC_PSERIES=n, or at least it can be depending on other config options and so on. Fix that by wrapping the entire content in an ifdef. Signed-off-by: -
Michael Ellerman authored
smp_query_cpu_stopped() and related #defines are currently in plpar_wrappers.h. The function actually does an RTAS call, not an hcall, and basically has nothing to do with plpar_wrappers.h Move it into pseries.h, where it can easily be used by the only two callers in pseries/smp.c and pseries/hotplug-cpu.c. Signed-off-by: -
Mathieu Malaterre authored
early_init() and machine_init() have no prototype, add one in asm-prototypes.h. Fixes the following warnings (treated as error in W=1): arch/powerpc/kernel/setup_32.c:68:30: error: no previous prototype for ‘early_init’ arch/powerpc/kernel/setup_32.c:99:21: error: no previous prototype for ‘machine_init’ Signed-off-by:
Mathieu Malaterre <malat@debian.org> [mpe: Move them to asm-prototypes.h, drop other functions] Signed-off-by:
-
Mathieu Malaterre authored
These functions can all be static, make it so. Signed-off-by:
-
Mathieu Malaterre authored
Rewrite function-like macro into regular static inline function to avoid a warning during macro expansion. Fix warning (treated as error in W=1): ./arch/powerpc/include/asm/uaccess.h:52:35: error: comparison of unsigned expression >= 0 is always true (((size) == 0) || (((size) - 1) <= ((segment).seg - (addr))))) ^ Suggested-by:Segher Boessenkool <segher@kernel.crashing.org> Signed-off-by:
Christophe Leroy <christophe.leroy@c-s.fr> Signed-off-by:
-
Mathieu Malaterre authored
Rewrite comparison since all values compared are of type `unsigned long`. Instead of using unsigned properties and rewriting the original code as: (originally suggested by Segher Boessenkool <segher@kernel.crashing.org>) #define pfn_valid(pfn) \ (((pfn) - ARCH_PFN_OFFSET) < (max_mapnr - ARCH_PFN_OFFSET)) Prefer a static inline function to make code as readable as possible. Fix a warning (treated as error in W=1): arch/powerpc/include/asm/page.h:129:32: error: comparison of unsigned expression >= 0 is always true [-Werror=type-limits] #define pfn_valid(pfn) ((pfn) >= ARCH_PFN_OFFSET && (pfn) < max_mapnr) ^ Suggested-by: -
Mathieu Malaterre authored
When neither CONFIG_ALTIVEC, nor CONFIG_VSX or CONFIG_PPC64 is defined, the array feature_properties is defined as an empty array, which in turn triggers the following warning (treated as error on W=1): arch/powerpc/kernel/prom.c: In function ‘check_cpu_feature_properties’: arch/powerpc/kernel/prom.c:298:16: error: comparison of unsigned expression < 0 is always false for (i = 0; i < ARRAY_SIZE(feature_properties); ++i, ++fp) { ^ Suggested-by: -
Mathieu Malaterre authored
Add missing prototypes for ppc_select() & ppc_fadvise64_64() to header asm-prototypes.h. Fix the following warnings (treated as errors in W=1) arch/powerpc/kernel/syscalls.c:87:1: error: no previous prototype for ‘ppc_select’ arch/powerpc/kernel/syscalls.c:119:6: error: no previous prototype for ‘ppc_fadvise64_64’ Signed-off-by:
-
