summaryrefslogtreecommitdiffstats
path: root/arch/mips/mm
Commit message (Collapse)AuthorAgeFilesLines
* [MIPS] tlb-r4k: Nuke broken paranoia error test.Ralf Baechle2008-07-301-4/+0
| | | | | | | | | Bug originally found and reported by Julia Lawall <julia@diku.dk>. I decieded that the whole error check was mostly useless paranoia and should be discarded. It would only ever trigger if r3k_have_wired_reg has a wrong value. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* mips: use generic show_mem()Johannes Weiner2008-07-262-38/+1
| | | | | | | | | | | | | | | | | | | Remove arch-specific show_mem() in favor of the generic version. This also removes the following redundant information display: - pages in swapcache, printed by show_swap_cache_info() where show_mem() calls show_free_areas(), which calls show_swap_cache_info(). And show_mem() does now actually print something on configurations with multiple nodes. Signed-off-by: Johannes Weiner <hannes@saeurebad.de> Acked-by: Ralf Baechle <ralf@linux-mips.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* dma-mapping: add the device argument to dma_mapping_error()FUJITA Tomonori2008-07-261-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add per-device dma_mapping_ops support for CONFIG_X86_64 as POWER architecture does: This enables us to cleanly fix the Calgary IOMMU issue that some devices are not behind the IOMMU (http://lkml.org/lkml/2008/5/8/423). I think that per-device dma_mapping_ops support would be also helpful for KVM people to support PCI passthrough but Andi thinks that this makes it difficult to support the PCI passthrough (see the above thread). So I CC'ed this to KVM camp. Comments are appreciated. A pointer to dma_mapping_ops to struct dev_archdata is added. If the pointer is non NULL, DMA operations in asm/dma-mapping.h use it. If it's NULL, the system-wide dma_ops pointer is used as before. If it's useful for KVM people, I plan to implement a mechanism to register a hook called when a new pci (or dma capable) device is created (it works with hot plugging). It enables IOMMUs to set up an appropriate dma_mapping_ops per device. The major obstacle is that dma_mapping_error doesn't take a pointer to the device unlike other DMA operations. So x86 can't have dma_mapping_ops per device. Note all the POWER IOMMUs use the same dma_mapping_error function so this is not a problem for POWER but x86 IOMMUs use different dma_mapping_error functions. The first patch adds the device argument to dma_mapping_error. The patch is trivial but large since it touches lots of drivers and dma-mapping.h in all the architecture. This patch: dma_mapping_error() doesn't take a pointer to the device unlike other DMA operations. So we can't have dma_mapping_ops per device. Note that POWER already has dma_mapping_ops per device but all the POWER IOMMUs use the same dma_mapping_error function. x86 IOMMUs use device argument. [akpm@linux-foundation.org: fix sge] [akpm@linux-foundation.org: fix svc_rdma] [akpm@linux-foundation.org: build fix] [akpm@linux-foundation.org: fix bnx2x] [akpm@linux-foundation.org: fix s2io] [akpm@linux-foundation.org: fix pasemi_mac] [akpm@linux-foundation.org: fix sdhci] [akpm@linux-foundation.org: build fix] [akpm@linux-foundation.org: fix sparc] [akpm@linux-foundation.org: fix ibmvscsi] Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Cc: Muli Ben-Yehuda <muli@il.ibm.com> Cc: Andi Kleen <andi@firstfloor.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@elte.hu> Cc: Avi Kivity <avi@qumranet.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* Merge branch 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linusLinus Torvalds2008-07-152-4/+2
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus: (54 commits) [MIPS] Remove mips_machtype for LASAT machines [MIPS] Remove mips_machtype from EMMA2RH machines [MIPS] Remove mips_machtype from ARC based machines [MIPS] MTX-1 flash partition setup move to platform devices registration [MIPS] TXx9: cleanup and fix some sparse warnings [MIPS] TXx9: rename asm-mips/mach-jmr3927 to asm-mips/mach-tx39xx [MIPS] remove machtype for group Toshiba [MIPS] separate rbtx4927_time_init() and rbtx4937_time_init() [MIPS] separate rbtx4927_arch_init() and rbtx4937_arch_init() [MIPS] txx9_cpu_clock setup move to rbtx4927_time_init() [MIPS] txx9_board_vec set directly without mips_machtype [MIPS] IP22: Add platform device for Indy volume buttons [MIPS] cmbvr4133: Remove support [MIPS] remove wrppmc_machine_power_off() [MIPS] replace inline assembler to cpu_wait() [MIPS] IP22/28: Add platform devices for HAL2 [MIPS] TXx9: Update and merge defconfigs [MIPS] TXx9: Make single kernel can support multiple boards [MIPS] TXx9: Update defconfigs [MIPS] TXx9: Reorganize PCI code ...
| * [MIPS] Make two functions staticDmitri Vorobiev2008-07-152-4/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The following routines uasm_rel_highest() uasm_rel_higher() are needlessly defined global. This patch makes them static. Compile-tested using a customized config for the Malta board. Booting the same board up to the shell prompt was also successful with this patch applied. Spotted by sparse. Signed-off-by: Dmitri Vorobiev <dmitri.vorobiev@movial.fi> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* | Merge branch 'generic-ipi' into generic-ipi-for-linusIngo Molnar2008-07-151-9/+9
|\ \ | |/ |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Conflicts: arch/powerpc/Kconfig arch/s390/kernel/time.c arch/x86/kernel/apic_32.c arch/x86/kernel/cpu/perfctr-watchdog.c arch/x86/kernel/i8259_64.c arch/x86/kernel/ldt.c arch/x86/kernel/nmi_64.c arch/x86/kernel/smpboot.c arch/x86/xen/smp.c include/asm-x86/hw_irq_32.h include/asm-x86/hw_irq_64.h include/asm-x86/mach-default/irq_vectors.h include/asm-x86/mach-voyager/irq_vectors.h include/asm-x86/smp.h kernel/Makefile Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * smp_call_function: get rid of the unused nonatomic/retry argumentJens Axboe2008-06-261-9/+9
| | | | | | | | | | | | | | | | It's never used and the comments refer to nonatomic and retry interchangably. So get rid of it. Acked-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
* | [MIPS] Fix 32bit kernels on R4k with 128 byte cache line sizeThomas Bogendoerfer2008-07-081-29/+32
| | | | | | | | | | | | | | | | | | | | | | | | The generated copy_page for R4k CPU with a 128 byte cache line size used Create Dirty Exclusive cache line operations even if only part of the cache line was filled. This change avoids generating cache operations, if only part of the cache line size is copied in one loop. It also increases the maxmimum loop size, because the generated code even fits into the available space for r4k CPUs with 128 byte cache line size. Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* | [MIPS] Atlas, decstation: Fix section mismatches triggered by defconfigsShane McDonald2008-07-082-5/+5
|/ | | | | | | | Resolve these mismatches by defining affected functions with the __cpuinit attribute, rather than __init. Signed-off-by: Shane McDonald <mcdonald.shane@gmail.com> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] Fix buggy use of kmap_coherent.Ralf Baechle2008-06-161-2/+5
| | | | | | | | Assuming the call of kmap_coherent in local_r4k_flush_cache_page doesn't need fixing this was skipped in fcae549295bcae801ac48fc1c2030ab8cc487020. Turns out it needed the same change after all. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] Fix the fix for divide by zero error in build_{clear,copy}_pageAtsushi Nemoto2008-06-161-1/+1
| | | | Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] Fix build for PNX platforms.Ralf Baechle2008-06-161-0/+26
| | | | | | Build error was caused by commit 351336929ccf222ae38ff0cb7a8dd5fd5c6236a0. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] Export empty_zero_page for sake of the ext4 module.Ralf Baechle2008-06-161-0/+1
| | | | Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* Fix divide by zero error in build_clear_page() and build_copy_page()Yoichi Yuasa2008-06-051-4/+6
| | | | | Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] R4700: Fix build_tlb_probe_entryThomas Bogendoerfer2008-06-051-1/+2
| | | | | | | | Treat R4700 like R4600 in build_tlb_probe_entry. Without this fix kernel will lock up. Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] unexport __kmap_atomic_to_pageAdrian Bunk2008-05-121-1/+0
| | | | | | | This patch removes the no longer used export of __kmap_atomic_to_page. Signed-off-by: Adrian Bunk <bunk@kernel.org> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] unexport copy_from_user_page()Dmitri Vorobiev2008-04-281-3/+0
| | | | | | | | | No users for the copy_from_user_page() routine exist outside of the core kernel code. Therefore, EXPORT_SYMBOL(copy_from_user_page) is useless, and this patch removes it. Signed-off-by: Dmitri Vorobiev <dmitri.vorobiev@gmail.com> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] unexport copy_to_user_page()Dmitri Vorobiev2008-04-281-2/+0
| | | | | | | | | The copy_to_user_page() function is called only in the core kernel code. Therefore, there is no need to export it. This patch removes EXPORT_SYMBOL(copy_to_user_page). Signed-off-by: Dmitri Vorobiev <dmitri.vorobiev@gmail.com> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] unexport copy_user_highpage()Dmitri Vorobiev2008-04-281-2/+0
| | | | | | | | | The copy_user_highpage() routine has no users outside of the core kernel code, so exporting this symbol is pointless. This patch removes EXPORT_SYMBOL(copy_user_highpage). Signed-off-by: Dmitri Vorobiev <dmitri.vorobiev@gmail.com> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] remove redundant display of free swap space in show_mem()Johannes Weiner2008-04-281-1/+0
| | | | | | Signed-off-by: Johannes Weiner <hannes@saeurebad.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] Add support for MIPS CMP platform.Ralf Baechle2008-04-282-7/+38
| | | | | | Signed-off-by: Chris Dearman <chris@mips.com> Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] Allow setting of the cache attribute at run time.Chris Dearman2008-04-282-9/+41
| | | | | | | | Slightly tacky, but there is a precedent in the sparc archirecture code. Signed-off-by: Chris Dearman <chris@mips.com> Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] All MIPS32 processors support64-bit physical addresses.Chris Dearman2008-04-282-2/+2
| | | | | | | Still, only the 4K may actually implement it. Signed-off-by: Chris Dearman <chris@mips.com> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] Reimplement clear_page/copy_pageThiemo Seufer2008-04-286-862/+725
| | | | | | | | | | | | | | | | | Fold the SB-1 specific implementation of clear_page/copy_page in the generic version, and rewrite that one in tlbex style. The immediate benefits: - It converts the compile-time workaround for SB-1 pass 1 prefetches to a more efficient run-time check. - It allows adjustment of loop unfolling, which helps to reduce the number of redundant cdex cache ops. - It fixes some esoteric cornercases (the cache line length calculations can go wrong, and support for 64k pages without prefetch instructions will overflow the addiu immediate). - Somewhat better guesses of "good" prefetch values. Signed-off-by: Thiemo Seufer <ths@networkno.de> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] Handle aliases in vmalloc correctly.Ralf Baechle2008-04-073-0/+32
| | | | | | | flush_cache_vmap / flush_cache_vunmap were calling flush_cache_all which - having been deprecated - turned into a nop ... Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] Add missing 4KEC TLB refill handlerThomas Bogendoerfer2008-04-011-0/+1
| | | | | | | | Early 4KEc were MIPS32r1 and therefore need some love to get a TLB refill handler. Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] Fix loads of section missmatchesRalf Baechle2008-03-1217-158/+148
| | | | Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] Fix typo in commentThiemo Seufer2008-03-121-1/+1
| | | | | | | We support now other page sizes as well. Signed-off-by: Thiemo Seufer <ths@networkno.de> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] SB1: Fix CONFIG_SIBYTE_DMA_PAGEOPS build failure.Thiemo Seufer2008-02-191-7/+7
| | | | | | | | | Fix type mismatch warnings for 64-bit kernel builds which trigger -Werror. The problem affects only SB-1 kernels with CONFIG_SIBYTE_DMA_PAGEOPS enabled. Signed-off-by: Thiemo Seufer <ths@networkno.de> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] Fix buggy invocations of kmap_coherent()Ralf Baechle2008-02-192-8/+16
| | | | | | | | kmap_coherent will only work correctly if the page it is called on is not marked dirty. If it's dirty the kernel address of the page should be used instead of a temporary mapping. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] Handle I-cache coherency in flush_cache_range()Ralf Baechle2008-02-191-3/+6
| | | | | | | | | | | | | | | | | | | | | | So far flush_cache_range() did't consider the I-cache largely because it did rarely ever matter to real world code. This was working primarily because normally code and data are don't share the same pages - with the exception of MIPS16 code which uses address constants embedded between the code. The following sequence of events may break the code: o MIPS16 executable being loaded o dynamic linker relocates the address constants embedded into the code: o Uses mprotect(2) to make code pages PROT_READ|PROT_WRITE o Performs the actual relocations by writing to the pages which likely are COW. Because no PROT_EXEC is set I-cache coherence will not be considered. o Uses mprotect(2) to switch code pages back to PROT_READ|PROT_EXEC. This results in a call to flush_cache_range() which also does not consider I-caches. o => executing the page just having been relocated may now result in the I-cache getting refilled with stale data from memory. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* arch/mips/: Spelling fixesJoe Perches2008-02-031-1/+1
| | | | | Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: Adrian Bunk <bunk@kernel.org>
* [MIPS] Split the micro-assembler from tlbex.c.Thiemo Seufer2008-02-014-960/+1111
| | | | | | | | | | | | | This patch moves the micro-assembler in a separate implementation, as it is useful for further run-time optimizations. The only change in behaviour is cutting down printk noise at kernel startup time. Checkpatch complains about macro parameters which aren't protected by parentheses. I believe this is a flaw in checkpatch, the paste operator used in those macros won't work with parenthesised parameters. Signed-off-by: Thiemo Seufer <ths@networkno.de> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] Alchemy: Au1210/Au1250 CPU supportManuel Lauss2008-01-292-0/+4
| | | | | | | | | | This patch adds IDs for new Au1200 variants: Au1210 and Au1250. They are essentially identical to the Au1200 except for the Au1210 which has a different SoC-ID in the PRId register [bits 31:24]. The Au1250 is a "Au1200 V0.2". Signed-off-by: Manuel Lauss <mano@roarinelk.homelinux.net> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] Use correct dma flushing in dma_cache_sync()Thomas Bogendoerfer2008-01-291-1/+1
| | | | | | | | Not cache coherent R10k systems (like IP28) need to do real cache invalidates in dma_cache_sync(). Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] Use real cache invalidateThomas Bogendoerfer2008-01-291-2/+2
| | | | | | | | | R10k non coherent machines need a real dma cache invalidate to get rid of speculative stores in cache. For other machines this promises a slight speedup. Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] tlbex.c: cleanup debug codeFranck Bui-Huu2008-01-291-57/+26
| | | | | Signed-off-by: Franck Bui-Huu <fbuihuu@gmail.com> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] tlbex.c: use __cacheline_aligned instead of __tlb_handler_align Franck Bui-Huu2008-01-291-6/+3
| | | | | Signed-off-by: Franck Bui-Huu <fbuihuu@gmail.com> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] tlbex.c: cleanup include filesFranck Bui-Huu2008-01-291-6/+0
| | | | | Signed-off-by: Franck Bui-Huu <fbuihuu@gmail.com> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] tlbex.c: Cleanup __init usages.Franck Bui-Huu2008-01-291-49/+49
| | | | | Signed-off-by: Franck Bui-Huu <fbuihuu@gmail.com> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] Remove useless S-cache flushes.Ralf Baechle2008-01-291-9/+0
| | | | Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] Use pte_present instead of open coded test for _PAGE_PRESENT.Ralf Baechle2008-01-291-1/+1
| | | | Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] R4000/R4400 daddiu erratum workaroundMaciej W. Rozycki2008-01-292-47/+61
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This complements the generic R4000/R4400 errata workaround code and adds bits for the daddiu problem. In most places it just modifies handwritten assembly code so that the assembler is allowed to use a temporary register as daddiu may now be treated as a macro that expands to a sequence of li and daddu. It is the AT register or, where AT is unavailable or used explicitly for another purpose, an explicitly-named register is selected, using the .set at=<reg> feature added recently to gas. This feature is only used if CONFIG_CPU_DADDI_WORKAROUNDS has been set, so if the workaround remains disabled, the required version of binutils stays unchanged. Similarly, daddiu instructions put in branch delay slots in noreorder fragments are now taken out of them and the assembler is allowed to reorder them itself as possible (which it does making the whole idea of scheduling them into delay slots manually questionable). Also in the very few places where such a simple conversion was not possible, a handcoded longer sequence is implemented. Other than that there are changes to code responsible for building the TLB fault and page clear/copy handlers to avoid daddiu as appropriate. These are only effective if the erratum is verified to be present at the run time. Finally there is a trivial update to __delay(), because it uses daddiu in a branch delay slot. Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] tlbex: Cleanup handling of R2 hazards in TLB handlers.Ralf Baechle2008-01-291-8/+6
| | | | Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] Wrong CONFIG option prevents setup of DMA zone.Thomas Bogendoerfer2008-01-111-1/+1
| | | | | Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] 64-bit Sibyte kernels need DMA32.Ralf Baechle2007-11-262-36/+44
| | | | | | | | | | | | Sibyte SOCs only have 32-bit PCI. Due to the sparse use of the address space only the first 1GB of memory is mapped at physical addresses below 1GB. If a system has more than 1GB of memory 32-bit DMA will not be able to reach all of it. For now this patch is good enough to keep Sibyte users happy but it seems eventually something like swiotlb will be needed for Sibyte. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] Sibyte: Replace use of removed IO_SPACE_BASE with IOADDR.Ralf Baechle2007-11-151-3/+3
| | | | Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] Convert reference to mem_map to pfn_to_page().Ralf Baechle2007-11-151-1/+1
| | | | | | This was crashing the combination of highmem and sparsemem. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] Sibyte: resurrect old cache hack.Ralf Baechle2007-11-151-1/+6
| | | | | | | | | | | | | | | | | | | The recent switch of the Sibyte SOCs from the processor specific cache managment code in c-sb1.c to c-r4k.c lost this old hack [MIPS] Hack for SB1 cache issues Removing flush_icache_page a while ago broke SB1 which was using an empty flush_data_cache_page function. This glues things well enough so a more efficient but also more intrusive solution can be found later. Signed-Off-By: Thiemo Seufer <ths@networkno.de> Signed-off-by: Ralf Baechle <ralf@linux-mips.org> in the hope it was no longer needed. As it turns it still is so resurrect it until there is a better solution. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
* [MIPS] MT: Fix bug in multithreaded kernels.Ralf Baechle2007-10-291-3/+18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When GDB writes a breakpoint into address area of inferior process the kernel needs to invalidate the modified memory in the inferior which is done by calling flush_cache_page which in turns calls r4k_flush_cache_page and local_r4k_flush_cache_page for VSMP or SMTC kernel via r4k_on_each_cpu(). As the VSMP and SMTC SMP kernels for 34K are running on a single shared caches it is possible to get away without interprocessor function calls. This optimization is implemented in r4k_on_each_cpu, so local_r4k_flush_cache_page is only ever called on the local CPU. This is where the following code in local_r4k_flush_cache_page() strikes: /* * If ownes no valid ASID yet, cannot possibly have gotten * this page into the cache. */ if (cpu_context(smp_processor_id(), mm) == 0) return; On VSMP and SMTC had a function of cpu_context() for each CPU(TC). So in case another CPU than the CPU executing local_r4k_cache_flush_page has not accessed the mm but one of the other CPUs has there may be data to be flushed in the cache yet local_r4k_cache_flush_page will falsely return leaving the I-cache inconsistent for the breakpoint. While the issue was discovered with GDB it also exists in local_r4k_flush_cache_range() and local_r4k_flush_cache(). Fixed by introducing a new function has_valid_asid which on MT kernels returns true if a mm is active on any processor in the system. This is relativly expensive since for memory acccesses in that loop cache misses have to be assumed but it seems the most viable solution for 2.6.23 and older -stable kernels. Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
OpenPOWER on IntegriCloud