- 14 Oct, 2018 32 commits
-
-
Nicholas Piggin authored
slb_flush_and_rebolt() is misleading, it is called in virtual mode, so it can not possibly change the stack, so it should not be touching the shadow area. And since vmalloc is no longer bolted, it should not change any bolted mappings at all. Change the name to slb_flush_and_restore_bolted(), and have it just load the kernel stack from what's currently in the shadow SLB area. Signed-off-by:
Nicholas Piggin <npiggin@gmail.com> Signed-off-by:
Michael Ellerman <mpe@ellerman.id.au>
-
Nicholas Piggin authored
When switching processes, currently all user SLBEs are cleared, and a few (exec_base, pc, and stack) are preloaded. In trivial testing with small apps, this tends to miss the heap and low 256MB segments, and it will also miss commonly accessed segments on large memory workloads. Add a simple round-robin preload cache that just inserts the last SLB miss into the head of the cache and preloads those at context switch time. Every 256 context switches, the oldest entry is removed from the cache to shrink the cache and require fewer slbmte if they are unused. Much more could go into this, including into the SLB entry reclaim side to track some LRU information etc, which would require a study of large memory workloads. But this is a simple thing we can do now that is an obvious win for common workloads. With the full series, process switching speed on the context_switch benchmark on POWER9/hash (with kernel speculation security masures disabled) increases from 140K/s to 178K/s (27%). POWER8 does not change much (within 1%), it's unclear why it does not see a big gain like POWER9. Booting to busybox init with 256MB segments has SLB misses go down from 945 to 69, and with 1T segments 900 to 21. These could almost all be eliminated by preloading a bit more carefully with ELF binary loading. Signed-off-by:
-
Nicholas Piggin authored
This will be used by the SLB code in the next patch, but for now this sets the slb_addr_limit to the correct size for 32-bit tasks. Signed-off-by:
-
Nicholas Piggin authored
Add 32-entry bitmaps to track the allocation status of the first 32 SLB entries, and whether they are user or kernel entries. These are used to allocate free SLB entries first, before resorting to the round robin allocator. Signed-off-by:
-
Nicholas Piggin authored
This patch moves SLB miss handlers completely to C, using the standard exception handler macros to set up the stack and branch to C. This can be done because the segment containing the kernel stack is always bolted, so accessing it with relocation on will not cause an SLB exception. Arbitrary kernel memory must not be accessed when handling kernel space SLB misses, so care should be taken there. However user SLB misses can access any kernel memory, which can be used to move some fields out of the paca (in later patches). User SLB misses could quite easily reconcile IRQs and set up a first class kernel environment and exit via ret_from_except, however that doesn't seem to be necessary at the moment, so we only do that if a bad fault is encountered. [ Credit to Aneesh for bug fixes, error checks, and improvements to bad address handling, etc ] Signed-off-by:
-
Nicholas Piggin authored
PPR is the odd register out when it comes to interrupt handling, it is saved in current->thread.ppr while all others are saved on the stack. The difficulty with this is that accessing thread.ppr can cause a SLB fault, but the SLB fault handler implementation in C change had assumed the normal exception entry handlers would not cause an SLB fault. Fix this by allocating room in the interrupt stack to save PPR. Signed-off-by:
-
Michael Ellerman authored
Now that we've split the user & kernel versions of pt_regs we need to be more careful in the ptrace code. For now we've ensured the location of the fields in both structs is the same, so most of the ptrace code doesn't need updating. But there are a few places where we use sizeof(pt_regs), and these will be wrong as soon as we increase the size of the kernel structure. So flip them all to use sizeof(user_pt_regs). Signed-off-by: -
Michael Ellerman authored
We use a shared definition for struct pt_regs in uapi/asm/ptrace.h. That means the layout of the structure is ABI, ie. we can't change it. That would be fine if it was only used to describe the user-visible register state of a process, but it's also the struct we use in the kernel to describe the registers saved in an interrupt frame. We'd like more flexibility in the content (and possibly layout) of the kernel version of the struct, but currently that's not possible. So split the definition into a user-visible definition which remains unchanged, and a kernel internal one. At the moment they're still identical, and we check that at build time. That's because we have code (in ptrace etc.) that assumes that they are the same. We will fix that code in future patches, and then we can break the strict symmetry between the two structs. Signed-off-by: -
Benjamin Herrenschmidt authored
It's never modified. Signed-off-by:
Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by:
-
Benjamin Herrenschmidt authored
It is never modified Signed-off-by:
-
Benjamin Herrenschmidt authored
It's not used anywhere else. Signed-off-by:
-
Christophe Leroy authored
In the same spirit as already done in pte query helpers, this patch changes pte setting helpers to perform endian conversions on the constants rather than on the pte value. In the meantime, it changes pte_access_permitted() to use pte helpers for the same reason. Signed-off-by:
Christophe Leroy <christophe.leroy@c-s.fr> Signed-off-by:
-
Christophe Leroy authored
_PAGE_PRIVILEGED corresponds to the SH bit which doesn't protect against user access but only disables ASID verification on kernel accesses. User access is controlled with _PMD_USER flag. Name it _PAGE_SH instead of _PAGE_PRIVILEGED _PAGE_HUGE corresponds to the SPS bit which doesn't really tells that's it is a huge page but only that it is not a 4k page. Name it _PAGE_SPS instead of _PAGE_HUGE Reviewed-by:
Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com> Signed-off-by:
-
Christophe Leroy authored
Do not include pte-common.h in nohash/32/pgtable.h As that was the last includer, get rid of pte-common.h Reviewed-by:
-
Christophe Leroy authored
Cache related flags like _PAGE_COHERENT and _PAGE_WRITETHRU are defined on most platforms. The platforms not defining them don't define any alternative. So we can give them a NUL value directly for those platforms directly. Reviewed-by:
-
Christophe Leroy authored
The 40xx defines _PAGE_HWWRITE while others don't. The 8xx defines _PAGE_RO instead of _PAGE_RW. The 8xx defines _PAGE_PRIVILEGED instead of _PAGE_USER. The 8xx defines _PAGE_HUGE and _PAGE_NA while others don't. Lets those platforms redefine pte_write(), pte_wrprotect() and pte_mkwrite() and get _PAGE_RO and _PAGE_HWWRITE off the common helpers. Lets the 8xx redefine pte_user(), pte_mkprivileged() and pte_mkuser() and get rid of _PAGE_PRIVILEGED and _PAGE_USER default values. Lets the 8xx redefine pte_mkhuge() and get rid of _PAGE_HUGE default value. Reviewed-by:
-
Christophe Leroy authored
nohash/64 only uses book3e PTE flags, so it doesn't need pte-common.h This also allows to drop PAGE_SAO and H_PAGE_4K_PFN from pte_common.h as they are only used by PPC64 Reviewed-by:
-
Christophe Leroy authored
The base kernel PAGE_XXXX definition sets are more or less platform specific. Lets distribute them close to platform _PAGE_XXX flags definition, and customise them to their exact platform flags. Also defines _PAGE_PSIZE and _PTE_NONE_MASK for each platform allthough they are defined as 0. Do the same with _PMD flags like _PMD_USER and _PMD_PRESENT_MASK Reviewed-by:
-
Christophe Leroy authored
Now the pte-common.h is only for nohash platforms, lets move pte_user() helper out of pte-common.h to put it together with other helpers. Reviewed-by:
-
Christophe Leroy authored
As done for book3s/64, add necessary flags/defines in book3s/32/pgtable.h and do not include pte-common.h It allows in the meantime to remove all related hash definitions from pte-common.h and to also remove _PAGE_EXEC default as _PAGE_EXEC is defined on all platforms except book3s/32. Reviewed-by:
-
Christophe Leroy authored
__P and __S flags are the same for all platform and should remain as is in the future, so avoid duplication. Reviewed-by:
-
Christophe Leroy authored
The following page flags in pte-common.h can be dropped: _PAGE_ENDIAN is only used in mm/fsl_booke_mmu.c and is defined in asm/nohash/32/pte-fsl-booke.h _PAGE_4K_PFN is nowhere defined nor used _PAGE_READ, _PAGE_WRITE and _PAGE_PTE are only defined and used in book3s/64 The following page flags in book3s/64/pgtable.h can be dropped as they are not used on this platform nor by common code. _PAGE_NA, _PAGE_RO, _PAGE_USER and _PAGE_PSIZE Reviewed-by:
-
Christophe Leroy authored
To reduce the complexity of flag_array, and allow the removal of default 0 value of non existing flags, lets have one flag_array table for each platform family with only the really existing flags. Reviewed-by:
-
Christophe Leroy authored
Get rid of platform specific _PAGE_XXXX in powerpc common code and use helpers instead. mm/dump_linuxpagetables.c will be handled separately Reviewed-by:
-
Christophe Leroy authored
The 'access' parameter of hash_preload() is either 0 or _PAGE_EXEC. Among the two versions of hash_preload(), only the PPC64 one is doing something with this 'access' parameter. In order to remove the use of _PAGE_EXEC outside platform code, 'access' parameter is replaced by 'is_exec' which will be either true of false, and the PPC64 version of hash_preload() creates the access flag based on 'is_exec'. Signed-off-by:
-
Christophe Leroy authored
In order to avoid using generic _PAGE_XXX flags in powerpc core functions, define helpers for all needed flags: - pte_mkuser() and pte_mkprivileged() to set/unset and/or unset/set _PAGE_USER and/or _PAGE_PRIVILEGED - pte_hashpte() to check if _PAGE_HASHPTE is set. - pte_ci() check if cache is inhibited (already existing on book3s/64) - pte_exprotect() to protect against execution - pte_exec() and pte_mkexec() to query and set page execution - pte_mkpte() to set _PAGE_PTE flag. - pte_hw_valid() to check _PAGE_PRESENT since pte_present does something different on book3s/64. On book3s/32 there is no exec protection, so pte_mkexec() and pte_exprotect() are nops and pte_exec() returns always true. Reviewed-by:
-
Christophe Leroy authored
In order to allow their use in nohash/32/pgtable.h, we have to move the following helpers in nohash/[32:64]/pgtable.h: - pte_mkwrite() - pte_mkdirty() - pte_mkyoung() - pte_wrprotect() Reviewed-by:
-
Christophe Leroy authored
book3s/32 doesn't define _PAGE_EXEC, so no need to use it. All other platforms define _PAGE_EXEC so no need to check it is not NUL when not book3s/32. Reviewed-by:
-
Christophe Leroy authored
In order to avoid multiple conversions, handover directly a pgprot_t to map_kernel_page() as already done for radix. Do the same for __ioremap_caller() and __ioremap_at(). Signed-off-by:
-
Christophe Leroy authored
Set PAGE_KERNEL directly in the caller and do not rely on a hack adding PAGE_KERNEL flags when _PAGE_PRESENT is not set. As already done for PPC64, use pgprot_cache() helpers instead of _PAGE_XXX flags in PPC32 ioremap() derived functions. Signed-off-by:
-
Christophe Leroy authored
In many places, ioremap_prot() and __ioremap() can be replaced with higher level functions like ioremap(), ioremap_coherent(), ioremap_cache(), ioremap_wc() ... Signed-off-by:
-
Christophe Leroy authored
ioremap_prot() with flag set to 0 relies on a hack in __ioremap_caller() which adds PAGE_KERNEL flags when the handed flags don't look like a valid set of flags (ie don't include _PAGE_PRESENT) The intention being to map cached memory, use ioremap_cache() instead. Signed-off-by:
-
- 13 Oct, 2018 8 commits
-
-
Christophe Leroy authored
_PAGE_WRITETHRU is a target specific flag. Prefer generic functions. Acked-by:
Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by:
-
Christophe Leroy authored
_PAGE_NO_CACHE is a platform specific flag. In addition, this flag is misleading because one would think it requests a noncached page whereas a noncached page is _PAGE_NO_CACHE | _PAGE_GUARDED _PAGE_NO_CACHE alone means write combined noncached page, so lets use ioremap_wc() instead. _PAGE_WRITETHRU is also platform specific flag. Use ioremap_wt() instead. Signed-off-by:
Daniel Vetter <daniel.vetter@ffwll.ch> Acked-by:
Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> Signed-off-by:
-
Christophe Leroy authored
Other arches have ioremap_wt() to map IO areas write-through. Implement it on PPC as well in order to avoid drivers using __ioremap(_PAGE_WRITETHRU) Also implement ioremap_coherent() to avoid drivers using __ioremap(_PAGE_COHERENT) Signed-off-by:
-
Gautham R. Shenoy authored
Live Partition Migrations require all the present CPUs to execute the H_JOIN call, and hence rtas_ibm_suspend_me() onlines any offline CPUs before initiating the migration for this purpose. The commit 85a88cab ("powerpc/pseries: Disable CPU hotplug across migrations") disables any CPU-hotplug operations once all the offline CPUs are brought online to prevent any further state change. Once the CPU-Hotplug operation is disabled, the code assumes that all the CPUs are online. However, there is a minor window in rtas_ibm_suspend_me() between onlining the offline CPUs and disabling CPU-Hotplug when a concurrent CPU-offline operations initiated by the userspace can succeed thereby nullifying the the aformentioned assumption. In this unlikely case these offlined CPUs will not call H_JOIN, resulting in a system hang. Fix this by verifying that all the present CPUs are actually online after CPU-Hotplug has been disabled, failing which we restore the state of the offline CPUs in rtas_ibm_suspend_me() and return an -EBUSY. Cc: Nathan Fontenot <nfont@linux.vnet.ibm.com> Cc: Tyrel Datwyler <tyreld@linux.vnet.ibm.com> Suggested-by:
Gautham R. Shenoy <ego@linux.vnet.ibm.com> Reviewed-by:
Nathan Fontenot <nfont@linux.vnet.ibm.com> Signed-off-by:
-
Gautham R. Shenoy authored
Currently on POWER9 SMT8 cores systems, in sysfs, we report the shared_cache_map for L1 caches (both data and instruction) to be the cpu-ids of the threads in SMT8 cores. This is incorrect since on POWER9 SMT8 cores there are two groups of threads, each of which shares its own L1 cache. This patch addresses this by reporting the shared_cpu_map correctly in sysfs for L1 caches. Before the patch /sys/devices/system/cpu/cpu0/cache/index0/shared_cpu_map : 000000ff /sys/devices/system/cpu/cpu0/cache/index1/shared_cpu_map : 000000ff /sys/devices/system/cpu/cpu1/cache/index0/shared_cpu_map : 000000ff /sys/devices/system/cpu/cpu1/cache/index1/shared_cpu_map : 000000ff After the patch /sys/devices/system/cpu/cpu0/cache/index0/shared_cpu_map : 00000055 /sys/devices/system/cpu/cpu0/cache/index1/shared_cpu_map : 00000055 /sys/devices/system/cpu/cpu1/cache/index0/shared_cpu_map : 000000aa /sys/devices/system/cpu/cpu1/cache/index1/shared_cpu_map : 000000aa Signed-off-by:
-
Gautham R. Shenoy authored
POWER9 SMT8 cores consist of two groups of threads, where threads in each group shares L1-cache. The scheduler is not aware of this distinction as the current sched-domain hierarchy has all the threads of the core defined at the SMT domain. SMT [Thread siblings of the SMT8 core] DIE [CPUs in the same die] NUMA [All the CPUs in the system] Due to this, we can observe run-to-run variance when we run a multi-threaded benchmark bound to a single core based on how the scheduler spreads the software threads across the two groups in the core. We fix this in this patch by defining each group of threads which share L1-cache to be the SMT level. The group of threads in the SMT8 core is defined to be the CACHE level. The sched-domain hierarchy after this patch will be : SMT [Thread siblings in the core that share L1 cache] CACHE [Thread siblings that are in the SMT8 core] DIE [CPUs in the same die] NUMA [All the CPUs in the system] Signed-off-by:
-
Gautham R. Shenoy authored
On IBM POWER9, the device tree exposes a property array identifed by "ibm,thread-groups" which will indicate which groups of threads share a particular set of resources. As of today we only have one form of grouping identifying the group of threads in the core that share the L1 cache, translation cache and instruction data flow. This patch adds helper functions to parse the contents of "ibm,thread-groups" and populate a per-cpu variable to cache information about siblings of each CPU that share the L1, traslation cache and instruction data-flow. It also defines a new global variable named "has_big_cores" which indicates if the cores on this configuration have multiple groups of threads that share L1 cache. For each online CPU, it maintains a cpu_smallcore_mask, which indicates the online siblings which share the L1-cache with it. Signed-off-by:
-
Michael Ellerman authored
If GCC is not built with glibc support then we must explicitly tell it which register to use for TLS mode stack protector, otherwise it will error out and the cc-option check will fail. Signed-off-by:
-
