- 02 3月, 2016 10 次提交
-
-
由 Cyril Bur 提交于
This patch adds the ability to be able to save the FPU registers to the thread struct without giving up (disabling the facility) next time the process returns to userspace. This patch optimises the thread copy path (as a result of a fork() or clone()) so that the parent thread can return to userspace with hot registers avoiding a possibly pointless reload of FPU register state. Signed-off-by: NCyril Bur <cyrilbur@gmail.com> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Cyril Bur 提交于
This prepares for the decoupling of saving {fpu,altivec,vsx} registers and marking {fpu,altivec,vsx} as being unused by a thread. Currently giveup_{fpu,altivec,vsx}() does both however optimisations to task switching can be made if these two operations are decoupled. save_all() will permit the saving of registers to thread structs and leave threads MSR with bits enabled. This patch introduces no functional change. Signed-off-by: NCyril Bur <cyrilbur@gmail.com> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Cyril Bur 提交于
Currently the FPU, VEC and VSX facilities are lazily loaded. This is not a problem unless a process is using these facilities. Modern versions of GCC are very good at automatically vectorising code, new and modernised workloads make use of floating point and vector facilities, even the kernel makes use of vectorised memcpy. All this combined greatly increases the cost of a syscall since the kernel uses the facilities sometimes even in syscall fast-path making it increasingly common for a thread to take an *_unavailable exception soon after a syscall, not to mention potentially taking all three. The obvious overcompensation to this problem is to simply always load all the facilities on every exit to userspace. Loading up all FPU, VEC and VSX registers every time can be expensive and if a workload does avoid using them, it should not be forced to incur this penalty. An 8bit counter is used to detect if the registers have been used in the past and the registers are always loaded until the value wraps to back to zero. Several versions of the assembly in entry_64.S were tested: 1. Always calling C. 2. Performing a common case check and then calling C. 3. A complex check in asm. After some benchmarking it was determined that avoiding C in the common case is a performance benefit (option 2). The full check in asm (option 3) greatly complicated that codepath for a negligible performance gain and the trade-off was deemed not worth it. Signed-off-by: NCyril Bur <cyrilbur@gmail.com> [mpe: Move load_vec in the struct to fill an existing hole, reword change log] Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au> fixup
-
由 Cyril Bur 提交于
Currently when threads get scheduled off they always giveup the FPU, Altivec (VMX) and Vector (VSX) units if they were using them. When they are scheduled back on a fault is then taken to enable each facility and load registers. As a result explicitly disabling FPU/VMX/VSX has not been necessary. Future changes and optimisations remove this mandatory giveup and fault which could cause calls such as clone() and fork() to copy threads and run them later with FPU/VMX/VSX enabled but no registers loaded. This patch starts the process of having MSR_{FP,VEC,VSX} mean that a threads registers are hot while not having MSR_{FP,VEC,VSX} means that the registers must be loaded. This allows for a smarter return to userspace. Signed-off-by: NCyril Bur <cyrilbur@gmail.com> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Cyril Bur 提交于
Load up the non volatile FPU and VMX regs and ensure that they are the expected value in a signal handler Signed-off-by: NCyril Bur <cyrilbur@gmail.com> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Cyril Bur 提交于
Loop in assembly checking the registers with many threads. Signed-off-by: NCyril Bur <cyrilbur@gmail.com> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Cyril Bur 提交于
Test that the non volatile floating point and Altivec registers get correctly preserved across the fork() syscall. fork() works nicely for this purpose, the registers should be the same for both parent and child Signed-off-by: NCyril Bur <cyrilbur@gmail.com> [mpe: Add include guards to basic_asm.h, minor formatting] Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Suraj Jitindar Singh 提交于
LTO can cause GCC to inline some functions which have attributes set. The act of inlining the functions can lead to GCC forgetting about the attributes which leads to incorrect tests. Notable example being: __attribute__((__target__("no-vsx"))) LTO can also interact strangely with custom assembly functions and cause tests to intermittently fail. Both these cases are hard to detect and require manual inspection of binaries which is unlikely to happen for all tests. Furthermore, LTO optimisations are not necessary for selftests and correctness is paramount and as such it is best to disable LTO. LTO can be enabled on a per test basis. A pseries_le_defconfig kernel on a POWER8 was used to determine that the same subset of selftests pass and fail with and without -flto in the common Makefile. Signed-off-by: NSuraj Jitindar Singh <sjitindarsingh@gmail.com> Reviewed-by: NCyril Bur <cyrilbur@gmail.com> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Michael Ellerman 提交于
Gcc helpfully points out that we're accessing past the end of the gprs array: tm-signal-msr-resv.c: In function 'signal_usr1': tm-signal-msr-resv.c:43:37: error: array subscript is above array bounds [-Werror=array-bounds] ucp->uc_mcontext.regs->gpr[PT_MSR] |= (7ULL); We haven't noticed previously because -flto was hiding it somehow. The code is confused, PT_MSR isn't a gpr, instead it's in uc_regs->gregs, so fix it. Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 David Gibson 提交于
htab_get_table_size() either retrieve the size of the hash page table (HPT) from the device tree - if the HPT size is determined by firmware - or uses a heuristic to determine a good size based on RAM size if the kernel is responsible for allocating the HPT. To support a PAPR extension allowing resizing of the HPT, we're going to want the memory size -> HPT size logic elsewhere, so split it out into a helper function. Signed-off-by: NDavid Gibson <david@gibson.dropbear.id.au> Reviewed-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
- 01 3月, 2016 8 次提交
-
-
由 David Gibson 提交于
This makes a number of cleanups to handling of mapping failures during memory hotplug on Power: For errors creating the linear mapping for the hot-added region: * This is now reported with EFAULT which is more appropriate than the previous EINVAL (the failure is unlikely to be related to the function's parameters) * An error in this path now prints a warning message, rather than just silently failing to add the extra memory. * Previously a failure here could result in the region being partially mapped. We now clean up any partial mapping before failing. For errors creating the vmemmap for the hot-added region: * This is now reported with EFAULT instead of causing a BUG() - this could happen for external reason (e.g. full hash table) so it's better to handle this non-fatally * An error message is also printed, so the failure won't be silent * As above a failure could cause a partially mapped region, we now clean this up. [mpe: move htab_remove_mapping() out of #ifdef CONFIG_MEMORY_HOTPLUG to enable this] Signed-off-by: NDavid Gibson <david@gibson.dropbear.id.au> Reviewed-by: NPaul Mackerras <paulus@samba.org> Reviewed-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 David Gibson 提交于
At the moment the hpte_removebolted callback in ppc_md returns void and will BUG_ON() if the hpte it's asked to remove doesn't exist in the first place. This is awkward for the case of cleaning up a mapping which was partially made before failing. So, we add a return value to hpte_removebolted, and have it return ENOENT in the case that the HPTE to remove didn't exist in the first place. In the (sole) caller, we propagate errors in hpte_removebolted to its caller to handle. However, we handle ENOENT specially, continuing to complete the unmapping over the specified range before returning the error to the caller. This means that htab_remove_mapping() will work sanely on a partially present mapping, removing any HPTEs which are present, while also returning ENOENT to its caller in case it's important there. There are two callers of htab_remove_mapping(): - In remove_section_mapping() we already WARN_ON() any error return, which is reasonable - in this case the mapping should be fully present - In vmemmap_remove_mapping() we BUG_ON() any error. We change that to just a WARN_ON() in the case of ENOENT, since failing to remove a mapping that wasn't there in the first place probably shouldn't be fatal. Signed-off-by: NDavid Gibson <david@gibson.dropbear.id.au> Reviewed-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 David Gibson 提交于
Currently, the only error that htab_remove_mapping() can report is -EINVAL, if removal of bolted HPTEs isn't implemeted for this platform. We make a few clean ups to the handling of this: * EINVAL isn't really the right code - there's nothing wrong with the function's arguments - use ENODEV instead * We were also printing a warning message, but that's a decision better left up to the callers, so remove it * One caller is vmemmap_remove_mapping(), which will just BUG_ON() on error, making the warning message redundant, so no change is needed there. * The other caller is remove_section_mapping(). This is called in the memory hot remove path at a point after vmemmap_remove_mapping() so if hpte_removebolted isn't implemented, we'd expect to have already BUG()ed anyway. Put a WARN_ON() here, in lieu of a printk() since this really shouldn't be happening. Signed-off-by: NDavid Gibson <david@gibson.dropbear.id.au> Reviewed-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Adam Buchbinder 提交于
Signed-off-by: NAdam Buchbinder <adam.buchbinder@gmail.com> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Luis Henriques 提交于
Instead of defining a local version of struct udphdr use the standard definition from <linux/udp.h>. The 'src' field is named 'source' in the <linux/udp.h> definition. Signed-off-by: NLuis Henriques <luis.henriques@canonical.com> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Luis Henriques 提交于
Instead of defining a local version of struct iphdr use the standard definition from <linux/ip.h>. Several fields in the <linux/ip.h> definition have different names: - proto -> protocol - src -> saddr - dest -> daddr - total_length -> tot_len - checksum -> check Also, 'ver_len' is composed by 'version' and 'ihl' in <linux/ip.h>. Signed-off-by: NLuis Henriques <luis.henriques@canonical.com> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Luis Henriques 提交于
Instead of defining the local struct vlantag use the standard definition of vlan_hdr from <linux/if_vlan.h>. The fields in the <linux/if_vlan.h> definition have different names: - vlan -> h_vlan_TCI - subtype -> h_vlan_encapsulated_proto While there, use also the ETH_P_IP macro instead of an hard-coded 0x0800 value. Signed-off-by: NLuis Henriques <luis.henriques@canonical.com> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Luis Henriques 提交于
Instead of defining a local version of struct ethhdr use the standard definition from <linux/if_ether.h>. The fields in the <linux/if_ether.h> definition have different names: - dest -> h_dest - src -> h_source - type -> h_proto While there, use a few other standard functions/macros: - eth_broadcast_addr (instead of a memset) - ETH_ALEN - ETH_P_8021Q Signed-off-by: NLuis Henriques <luis.henriques@canonical.com> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
- 29 2月, 2016 7 次提交
-
-
由 Paul Mackerras 提交于
Now that other PTE fields have been moved out of the way, we can expand the RPN field of the PTE on 64-bit Book 3S systems and align it with the RPN field in the radix PTE format used by PowerISA v3.0 CPUs in radix mode. For 64k page size, this means we need to move the _PAGE_COMBO and _PAGE_4K_PFN bits. Signed-off-by: NPaul Mackerras <paulus@samba.org> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Paul Mackerras 提交于
This moves the _PAGE_SPECIAL and _PAGE_SOFT_DIRTY bits in the Linux PTE on 64-bit Book 3S systems to bit positions which are designated for software use in the radix PTE format used by PowerISA v3.0 CPUs in radix mode. Signed-off-by: NPaul Mackerras <paulus@samba.org> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Paul Mackerras 提交于
This moves the _PAGE_EXEC, _PAGE_RW and _PAGE_USER bits around in the Linux PTE on 64-bit Book 3S systems to correspond with the bit positions used in radix mode by PowerISA v3.0 CPUs. This also adds a _PAGE_READ bit corresponding to the read permission bit in the radix PTE. _PAGE_READ is currently unused but could possibly be used in future to improve pte_protnone(). Signed-off-by: NPaul Mackerras <paulus@samba.org> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Paul Mackerras 提交于
This moves the _PAGE_HASHPTE, _PAGE_F_GIX and _PAGE_F_SECOND fields in the Linux PTE on 64-bit Book 3S systems to the most significant byte. Of the 5 bits, one is a software-use bit and the other four are reserved bit positions in the PowerISA v3.0 radix PTE format. Using these bits is OK because these bits are all to do with tracking the HPTE(s) associated with the Linux PTE, and therefore won't be needed in radix mode. This frees up bit positions in the lower two bytes. Signed-off-by: NPaul Mackerras <paulus@samba.org> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Paul Mackerras 提交于
This changes _PAGE_PTE for 64-bit Book 3S processors from 0x1 to 0x4000_0000_0000_0000, because that bit is used as the L (leaf) bit by PowerISA v3.0 CPUs in radix mode. The "leaf" bit indicates that the PTE points to a page directly rather than another radix level, which is what the _PAGE_PTE bit means. Signed-off-by: NPaul Mackerras <paulus@samba.org> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Paul Mackerras 提交于
This changes _PAGE_PRESENT for 64-bit Book 3S processors from 0x2 to 0x8000_0000_0000_0000, because that is where PowerISA v3.0 CPUs in radix mode will expect to find it. Signed-off-by: NPaul Mackerras <paulus@samba.org> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Paul Mackerras 提交于
This changes the Linux page tables to store physical addresses rather than kernel virtual addresses in the upper levels of the tree (pgd, pud and pmd) for 64-bit Book 3S machines. This also changes the hugepd pointers used to implement hugepages when the base page size is 4k to store physical addresses rather than virtual addresses (again just for 64-bit Book3S machines). This frees up some high order bits, and will be needed with PowerISA v3.0 machines which read the page table tree in hardware in radix mode. Signed-off-by: NPaul Mackerras <paulus@samba.org> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
- 27 2月, 2016 2 次提交
-
-
由 Paul Mackerras 提交于
This frees up bits 57-63 in the Linux PTE on 64-bit Book 3S machines. In the 4k page case, this is done just by reducing the size of the RPN field to 39 bits, giving 51-bit real addresses. In the 64k page case, we had 10 unused bits in the middle of the PTE, so this moves the RPN field down 10 bits to make use of those unused bits. This means the RPN field is now 3 bits larger at 37 bits, giving 53-bit real addresses in the normal case, or 49-bit real addresses for the special 4k PFN case. We are doing this in order to be able to move some other PTE bits into the positions where PowerISA V3.0 processors will expect to find them in radix-tree mode. Ultimately we will be able to move the RPN field to lower bit positions and make it larger. Signed-off-by: NPaul Mackerras <paulus@samba.org> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Paul Mackerras 提交于
No code changes. Signed-off-by: NPaul Mackerras <paulus@samba.org> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
- 25 2月, 2016 1 次提交
-
-
由 Michael Ellerman 提交于
Pull in our current fixes from 4.5, in particular the "Fix Multi hit ERAT" bug is causing folks some grief when testing next.
-
- 24 2月, 2016 5 次提交
-
-
由 Balbir Singh 提交于
I ran into this issue while debugging an early boot problem. The system hit a BUG_ON() but report bug failed to print the line number and file name. The reason being that the system was running in real mode and report_bug() searches for addresses in the PAGE_OFFSET+ region. Suggested-by: NPaul Mackerras <paulus@samba.org> Signed-off-by: NBalbir Singh <bsingharora@gmail.com> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 pan xinhui 提交于
__xchg_called_with_bad_pointer() can't tell us which code uses {cmp}xchg with an unsupported size, and no error is reported until the link stage. To make such problems easier to debug, use BUILD_BUG_ON_MSG() instead. Signed-off-by: Npan xinhui <xinhui.pan@linux.vnet.ibm.com> [mpe: Tweak change log wording & add relaxed/acquire] Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au> fixup
-
由 Jeremy Kerr 提交于
Most current OpenPOWER platforms have an AST BMC, so add graphics support via the AST DRM driver. Signed-off-by: NJeremy Kerr <jk@ozlabs.org> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Jeremy Kerr 提交于
There are a few firmware-provided interfaces for OpenPOWER platforms: the PRD infrastructure, IPMI support, and MTD access to the PNOR flash. This change adds these to powernv_defconfig Signed-off-by: NJeremy Kerr <jk@ozlabs.org> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Jeremy Kerr 提交于
This change adds a defconfig for the non-virtualised power platforms, based on pseries_defconfig, but without pseries, and little-endian, and no OF trampoline. Signed-off-by: NJeremy Kerr <jk@ozlabs.org> Acked-by: NJoel Stanley <joel@jms.id.au> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
- 22 2月, 2016 6 次提交
-
-
由 Michael Neuling 提交于
Add a cputable entry for POWER9. More code is required to actually boot and run on a POWER9 but this gets the base piece in which we can start building on. Copies over from POWER8 except for: - Adds a new CPU_FTR_ARCH_300 bit to start hanging new architecture features from (in subsequent patches). - Advertises new user features bits PPC_FEATURE2_ARCH_3_00 & HAS_IEEE128 when on POWER9. - Drops CPU_FTR_SUBCORE. - Drops PMU code and machine check. Signed-off-by: NMichael Neuling <mikey@neuling.org> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Michael Neuling 提交于
Use defines for literals __init_tlb_power[78] rather than hand coding them. Signed-off-by: NMichael Neuling <mikey@neuling.org> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Michael Neuling 提交于
Subcores isn't really part of the 2.07 architecture but currently we turn it on using the 2.07 feature bit. Subcores is really a POWER8 specific feature. This adds a new CPU_FTR bit just for subcores and moves the subcore init code over to use this. Reviewed-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: NMichael Neuling <mikey@neuling.org> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Andrew Donnellan 提交于
When initialising OPAL interfaces, there is a possibility that opal_msglog_init() may fail to initialise the msglog/memory console. Fix opal_msglog_sysfs_init() so it doesn't try to create sysfs entry for the msglog if this occurs. Suggested-by: NJoel Stanley <joel@jms.id.au> Fixes: 9b4fffa1 ("powerpc/powernv: new function to access OPAL msglog") Signed-off-by: NAndrew Donnellan <andrew.donnellan@au1.ibm.com> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Aneesh Kumar K.V 提交于
We can get a hash pte fault with 4k base page size and find the pte already inserted with 64K base page size. In that case we need to clear the existing slot information from the old pte. Fix this correctly With THP, we also clear the slot information with respect to all the 64K hash pte mapping that 16MB page. They are all invalid now. This make sure we don't find the slot valid when we fault with 4k base page size. Finding the slot valid should not result in any wrong behavior because we do check again in hash page table for the validity. But we can avoid that check completely. Fixes: a43c0eb8 ("powerpc/mm: Convert 4k hash insert to C") Signed-off-by: NAneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
由 Gavin Shan 提交于
During error recovery, the device could be removed as part of the partial hotplug. The criterion used to come with partial hotplug is: if the device driver provides error_detected(), slot_reset() and resume() callbacks, it's immune from hotplug. Otherwise, it's going to experience partial hotplug during EEH recovery. But the criterion isn't correct enough: mlx4_core driver for Mellanox adapters provides error_detected(), slot_reset() callbacks, but resume() isn't there. Those Mellanox adapters won't be to involved in the partial hotplug. This fixes the criterion to a practical one: adpater with driver that provides error_detected(), slot_reset() will be immune from partial hotplug. resume() isn't mandatory. Fixes: f2da4ccf ("powerpc/eeh: More relaxed hotplug criterion") Cc: stable@vger.kernel.org #v4.4+ Signed-off-by: NGavin Shan <gwshan@linux.vnet.ibm.com> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-
- 17 2月, 2016 1 次提交
-
-
由 Boqun Feng 提交于
Implement cmpxchg{,64}_relaxed and atomic{,64}_cmpxchg_relaxed, based on which _release variants can be built. To avoid superfluous barriers in _acquire variants, we implement these operations with assembly code rather use __atomic_op_acquire() to build them automatically. For the same reason, we keep the assembly implementation of fully ordered cmpxchg operations. However, we don't do the similar for _release, because that will require putting barriers in the middle of ll/sc loops, which is probably a bad idea. Note cmpxchg{,64}_relaxed and atomic{,64}_cmpxchg_relaxed are not compiler barriers. Signed-off-by: NBoqun Feng <boqun.feng@gmail.com> Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>
-