aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAge
* IB/ipath: Set CPU affinity earlyBryan O'Sullivan2006-09-28
| | | | | | | | | This change moves around port assignment so that it happens before any memory is allocated. This allows memory to be allocated on an appropriate CPU, which improves performance for users of /dev/ipath. Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: Fix EEPROM read when driver is compiled with -OsBryan O'Sullivan2006-09-28
| | | | | | | | | The EEPROM is read via programmable I/O pins. When the driver is compiled -Os, the CPU can speculatively read the I/O value before it is valid. This patch fixes the problem. Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: Fix and recover TXE piobuf and PBC parity errorsBryan O'Sullivan2006-09-28
| | | | | | | | | | | | | | | | We can sometimes trigger parity errors due to processor speculative reads to our write-combined memory (mostly seen on Woodcrest). Add a stats counter for these. Factored out the sendbuffererror buffer cancellation code so it can be used in the new handling; suppress likely subsequent error messages if within two jiffies of the cancellation. Also restore 2 dropped TXE lines on hwe_bitsextant noticed while debugging. Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: Change HT CRC message to indicate how to resolve problemBryan O'Sullivan2006-09-28
| | | | | | | | The system must be powercycled to clear a HT CRC error; reloading the driver is not enough. Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: Clean up module exit codeBryan O'Sullivan2006-09-28
| | | | | Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: Call mtrr_del with correct argumentsBryan O'Sullivan2006-09-28
| | | | | | | | We were passing 0 for base and length, which worked on older kernels, but it doesn't seem to any longer. Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: Flush RWQEs if access error or invalid error seenBryan O'Sullivan2006-09-28
| | | | | | | | If the receiver goes into the error state, we need to flush the posted receive WQEs. Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: Improved support for PowerPCBryan O'Sullivan2006-09-28
| | | | | Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: Drop unnecessary "(void *)" castsBryan O'Sullivan2006-09-28
| | | | | Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: Support multiple simultaneous devices of different typesBryan O'Sullivan2006-09-28
| | | | | | | | Prior to this change, the driver was not able to support a HT and PCIE card simultaneously present in the same machine. Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: Fix mismatch in shifts and masks for printing debug infoBryan O'Sullivan2006-09-28
| | | | | | | | | | Fixed mismatch in linkstate/trainingstate shifts and masks in the IPATH_IBSTATE_MASK macro. It kept some linktrainingstates from being printed correctly in debug; no functionality issue unless I misread the code. Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: Fix compiler warnings and errors on non-x86_64 systemsBryan O'Sullivan2006-09-28
| | | | | Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: Print more informative parity error messagesBryan O'Sullivan2006-09-28
| | | | | Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: Ensure that PD of MR matches PD of QP checking the RkeyBryan O'Sullivan2006-09-28
| | | | | Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: RC and UC should validate SLID and DLIDBryan O'Sullivan2006-09-28
| | | | | | | This is required for IB conformance (spec ch. 9.6.1.5). Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: Only allow complete writes to flashBryan O'Sullivan2006-09-28
| | | | | | | | Don't allow a write to the eeprom from ipathfs unless the write is exactly 128 bytes and starts at offset 0. Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: Count SRQs properlyBryan O'Sullivan2006-09-28
| | | | | Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: Lock and count allocated CQs properlyBryan O'Sullivan2006-09-28
| | | | | Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: Clean up handling of GUID 0Bryan O'Sullivan2006-09-28
| | | | | | | | Respond with an error to the SM if our GUID is 0, and don't allow the user to set our GUID to 0. Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: Unregister from IB core earlyBryan O'Sullivan2006-09-28
| | | | | | | | This gives upper-level protocols a chance to unregister while the device is still usable. Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: Support revision 2 InfiniPath PCIE devicesBryan O'Sullivan2006-09-28
| | | | | | | This also entailed a little GPIO-interrupt general cleanup. Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: Driver support for userspace sharing of HW contextsBryan O'Sullivan2006-09-28
| | | | | | | | | This allows multiple userspace processes to share a single hardware context in a master/slave arrangement. It is backwards binary compatible with existing userspace. Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: Fix memory leak if allocation failsBryan O'Sullivan2006-09-28
| | | | | | | | If the second allocation failed, the first structure allocated in this routine was not freed. Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/ipath: Limit # of packets sent without an ACK receivedBryan O'Sullivan2006-09-28
| | | | | | | | The sender requests an ACK every 1/2 MB to avoid retransmit timeouts that were causing MVAPICH mod_bw to fail after a predictable number of sends. Signed-off-by: Bryan O'Sullivan <bryan.osullivan@qlogic.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/iser: Fix the description of iSER in KconfigErez Zilber2006-09-28
| | | | | | | Fix the description of iSER in Kconfig. It is not accurate. Signed-off-by: Erez Zilber <erezz@voltaire.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/iser: DMA unmap unaligned for RDMA data before touching itErez Zilber2006-09-28
| | | | | | | | | | | | iSER uses the DMA mapping api to map the page holding the SCSI command data to the HCA DMA address space. When the command data is not aligned for RDMA, the data is copied to/from an allocated buffer which in turn is used for executing this command. The pages associated with the command must be unmapped before being touched. Signed-off-by: Erez Zilber <erezz@voltaire.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* IB/iser: Have iSER data transaction object point to iSER connErez Zilber2006-09-28
| | | | | | | | | | | | | iSER uses a data transaction object (struct iser_dto) as part of its IB data descriptors (struct iser_desc) management. It also uses a hierarchy of connection structures pointing to each other. A DTO may exist even after the iscsi_iser connection pointed by it is destroyed (eg one that is bound to a post receive buffer which was flushed by the IB HW). Hence DTOs need point to the lowest connection, which is struct iser_conn. Signed-off-by: Erez Zilber <erezz@voltaire.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* RDMA/amso1100: Fix memory leak in c2_reg_phys_mr()Roland Dreier2006-09-28
| | | | | | | | | If the allocation of mr fails, then c2_reg_phys_mr() leaks the page_list array it allocated earlier. This was Coverity CID #1413. Signed-off-by: Roland Dreier <rolandd@cisco.com>
* RDMA/amso1100: Fix error path in c2_llp_accept()Eric Sesterhenn2006-09-28
| | | | | | | | | | | Another NULL dereference spotted by the Coverity checker (cid #1395): In case we can't alloc the vq_req, we goto bail1, where we call vq_req_free(c2dev, vq_req); which then dereferences vq_req. Signed-off-by: Eric Sesterhenn <snakebyte@gmx.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Acked-by: Tom Tucker <tom@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* RDMA/amso1100: Fix compile warningsRoland Dreier2006-09-27
| | | | | | | Make sure all 64-bit quantities are cast to unsigned long long when printed with "%ll" printk formats. Signed-off-by: Roland Dreier <rolandd@cisco.com>
* Merge branch 'release' of ↵Linus Torvalds2006-09-27
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6 * 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6: [IA64] minor reformatting to vmlinux.lds.S [IA64] CMC/CPE: Reverse the order of fetching log and checking poll threshold [IA64] PAL calls need physical mode, stacked [IA64] ar.fpsr not set on MCA/INIT kernel entry [IA64] printing support for MCA/INIT [IA64] trim output of show_mem() [IA64] show_mem() printk levels [IA64] Make gp value point to Region 5 in mca handler Revert "[IA64] Unwire set/get_robust_list" [IA64] Implement futex primitives [IA64-SGI] Do not request DMA memory for BTE [IA64] Move perfmon tables from thread_struct to pfm_context [IA64] Add interface so modules can discover whether multithreading is on. [IA64] kprobes: fixup the pagefault exception caused by probehandlers [IA64] kprobe opcode 16 bytes alignment on IA64 [IA64] esi-support [IA64] Add "model name" to /proc/cpuinfo
| * [IA64] minor reformatting to vmlinux.lds.SAl Stone2006-09-26
| | | | | | | | | | | | | | | | Minor reformatting to vmlinux.lds.S to make it 80-column usable, in accordance with Linux coding style. Signed-off-by: Al Stone <ahs3@fc.hp.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
| * [IA64] CMC/CPE: Reverse the order of fetching log and checking poll thresholdHidetoshi Seto2006-09-26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch reverses the order of fetching log from SAL and checking poll threshold. This will fix following trivial issues: - If SAL_GET_SATE_INFO is unbelievably slow (due to huge system or just its silly implementation) and if it takes more than 1/5 sec, CMCI/CPEI will never switch to CMCP/CPEP. - Assuming terrible flood of interrupt (continuous corrected errors let all CPUs enter to handler at once and bind them in it), CPUs will be serialized by IA64_LOG_LOCK(*). Now we check the poll threshold after the lock and log fetch, so we need to call SAL_GET_STATE_INFO (num_online_cpus() + 4) times in the worst case. if we can check the threshold before the lock, we can shut up interrupts quickly without waiting preceding log fetches, and the number of times will be reduced to (num_online_cpus()) in the same situation. Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
| * [IA64] PAL calls need physical mode, stackedRuss Anderson2006-09-26
| | | | | | | | | | | | | | | | PAL_CACHE_READ and PAL_CACHE_WRITE need to be called in physical mode with stacked registers. Signed-off-by: Russ Anderson (rja@sgi.com) Signed-off-by: Tony Luck <tony.luck@intel.com>
| * [IA64] ar.fpsr not set on MCA/INIT kernel entryRuss Anderson2006-09-26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When entering the kernel due to an MCA or INIT, ar.fpsr (ar40) was not getting set to the kernel default value (remaining at the user value). The effect depends on the user setting of ar.fpsr. In the test case, the effect was addresses printing with strange hex values. Setting ar.fpsr in ia64_set_kernel_registers sets it for both the MCA and INIT paths. The user value of ar.fpsr is correctly saved (in ia64_state_save) and restored (in ia64_state_restore). Below is an example of output with very strange hex values. Anyone know the value of hex 'g'? :-) Processes interrupted by INIT - 0 (cpu 14 task 0xdfffg55g7a4c6gA) Signed-off-by: Russ Anderson (rja@sgi.com) Signed-off-by: Tony Luck <tony.luck@intel.com>
| * [IA64] printing support for MCA/INITHidetoshi Seto2006-09-26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Printing message to console from MCA/INIT handler is useful, however doing oops_in_progress = 1 in them exactly makes something in kernel wrong. Especially it sounds ugly if system goes wrong after returning from recoverable MCA. This patch adds ia64_mca_printk() function that collects messages into temporary-not-so-large message buffer during in MCA/INIT environment and print them out later, after returning to normal context or when handlers determine to down the system. Also this print function is exported for use in extensional MCA handler. It would be useful to describe detail about recovery. NOTE: I don't think it is sane thing if temporary message buffer is enlarged enough to hold whole stack dumps from INIT, so buffering is disabled during stack dump from INIT-monarch (= default_monarch_init_process). please fix it in future. Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com> Acked-by: Russ Anderson <rja@sgi.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
| * [IA64] trim output of show_mem()Jes Sorensen2006-09-26
| | | | | | | | | | | | | | | | Cut the number of lines of memory info output per node from five to one line. Signed-off-by: Jes Sorensen <jes@sgi.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
| * [IA64] show_mem() printk levelsJes Sorensen2006-09-26
| | | | | | | | | | | | | | | | | | Use the default sysrq printk level for printing show_mem() output both for disconfig and contig versions. This is consistent with the printk level used on other architectures (well ia32 at least). Signed-off-by: Jes Sorensen <jes@sgi.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
| * [IA64] Make gp value point to Region 5 in mca handlerZou Nan hai2006-09-26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | MCA dispatch code take physical address of GP passed from SAL, then call DATA_PA_TO_VA twice on GP before call into C code. The first time is in ia64_set_kernel_register, the second time is in VIRTUAL_MODE_ENTER. The gp is changed to a virtual address in region 7 because DATA_PA_TO_VA is implemented by dep instruction. However when notify blocks were called from MCA handler code, because notify blocks are supported by callback function pointers, gp value value was switched to region 5 again. The patch set gp register to kernel gp of region 5 at entry of MCA dispatch. Signed-off-by: Zou Nan hai <nanhai.zou@intel.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
| * Revert "[IA64] Unwire set/get_robust_list"Tony Luck2006-09-26
| | | | | | | | | | | | | | | | | | This reverts commit 2636255488484e04d6d54303d2b0ec30f7ef7e02. Jakub Jelinek provided the missing futex_atomic_cmpxchg_inatomic() function, so now it should be safe to re-enable these syscalls. Signed-off-by: Tony Luck <tony.luck@intel.com>
| * [IA64] Implement futex primitivesJakub Jelinek2006-09-26
| | | | | | | | | | | | | | | | | | Implement futex_atomic_op_inuser() and futex_atomic_cmpxchg_inatomic() on IA64 in order to fully support all futex functionality. Signed-off-by: Jakub Jelinek <jakub@redhat.com> Signed-off-by: David Woodhouse <dwmw2@infradead.org> Signed-off-by: Tony Luck <tony.luck@intel.com>
| * [IA64-SGI] Do not request DMA memory for BTEChristoph Lameter2006-09-26
| | | | | | | | | | | | | | | | | | | | | | | | The GFP_DMA option usually does nothing on SN2 since all of memory is in thei DMA zone and the BTE has always been capable of addressing all of memory. So there is no need to get memory from a restricted range of memory (which is what GFP_DMA is for). Remove useless __GFP_DMA option. Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
| * [IA64] Move perfmon tables from thread_struct to pfm_contextKeshavamurthy Anil S2006-09-26
| | | | | | | | | | | | | | | | | | | | | | This patch renders thread_struct->pmcs[] and thread_struct->pmds[] OBSOLETE. The actual table is moved to pfm_context structure which saves space in thread_struct (in turn saving space in task_struct which frees up more space for kernel stacks). Signed-off-by: Stephane Eranian <eranian@hpl.hp.com> Signed-off-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
| * [IA64] Add interface so modules can discover whether multithreading is on.Stephane Eranian2006-09-26
| | | | | | | | | | | | | | | | Add is_multithreading_enabled() to check whether multi-threading is enabled independently of which cpu is currently online Signed-off-by: stephane eranian <eranian@hpl.hp.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
| * [IA64] kprobes: fixup the pagefault exception caused by probehandlersKeshavamurthy Anil S2006-09-26
| | | | | | | | | | | | | | | | | | | | If the user-specified kprobe handler causes the page fault when accessing user space address, fixup this fault since do_page_fault() should not continue as the kprobe handler are run with preemption disabled. Signed-off-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Tony Luck <tony.luck@intel.com>
| * [IA64] kprobe opcode 16 bytes alignment on IA64bibo mao2006-09-26
| | | | | | | | | | | | | | | | | | | | | | | | On IA64 instruction opcode must be 16 bytes alignment, in kprobe structure there is one element to save original instruction, currently saved opcode is not statically allocated in kprobe structure, that can not assure 16 bytes alignment. This patch dynamically allocated kprobe instruction opcode to assure 16 bytes alignment. Signed-off-by: bibo mao <bibo.mao@intel.com> Acked-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
| * Pull esi-support into release branchTony Luck2006-09-26
| |\
| | * [IA64] esi-supportDavid Mosberger-Tang2006-06-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add support for making ESI calls [1]. ESI stands for "Extensible SAL specification" and is basically a way for invoking firmware subroutines which are identified by a GUID. I don't know whether ESI is used by vendors other than HP (if you do, please let me know) but as firmware "backdoors" go, this seems one of the cleaner methods, so it seems reasonable to support it, even though I'm not aware of any publicly documented ESI calls. I'd have liked to make the ESI module completely stand-alone, but unfortunately that is not easily (or not at all) possible because in order to make ESI calls in physical mode, a small stub similar to the EFI stub is needed in the kernel proper. I did try to create a stub that would work in user-level, but it quickly got ugly beyond recognition (e.g., the stub had to make assumptions about how the module-loader generated call-stubs work) and I didn't even get it to work (that's probably fixable, but I didn't bother because I concluded it was too ugly anyhow). While it's not terribly elegant to have kernel code which isn't actively used in the kernel proper, I think it might be worth making an exception here for two reasons: the code is trivially small (all that's really needed is esi_stub.S) and by including it in the normal kernel distro, it might encourage other OEMs to also use ESI, which I think would be far better than each inventing their own firmware "backdoor". The code was originally written by Alex. I just massaged and packaged it a bit (and perhaps messed up some things along the way...). Changes since first version of patch that was posted to mailing list: * Export ia64_esi_call and ia64_esi_call_phys() as GPL symbols. * Disallow building esi.c as a module for now. Building as a module would currently lead to an unresolved reference to "sal_lock" on SMP kernels because that symbol doesn't get exported. * Export esi_call_phys() only if ESI is enabled. * Remove internal stuff from esi.h and add a "proc_type" argument to ia64_esi_call() such that serialization-requirements can be expressed (ESI follows SAL here, where procedure calls may have to be serialized, are MP-safe, or MP-safe andr reentrant). [1] h21007.www2.hp.com/dspp/tech/tech_TechDocumentDetailPage_IDX/1,1701,919,00.html Signed-off-by: David Mosberger <David.Mosberger@acm.org> Signed-off-by: Alex Williamson <alex.williamson@hp.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
| * | Pull model-name into release branchTony Luck2006-09-26
| |\ \
| | * | [IA64] Add "model name" to /proc/cpuinfoTony Luck2006-06-05
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Linux ia64 port tried to decode the processor family number to something human-readable, but Intel brandnames don't change synchronously with updates to the family number. Adopt a more i386-like approach and just print the family number in decimal. Add a new field "model name" that uses PAL_BRAND_INFO to find the official name for the cpu, or on older systems, falls back to using the well-known codenames (Merced, McKinley, Madison). Signed-off-by: Tony Luck <tony.luck@intel.com>
generated by cgit v1.2.2 (git 2.25.0) at 2025-01-24 08:12:06 -0500