UbiquitousOS
/
OpenCloudOS-Kernel
mirror of https://gitee.com/OpenCloudOS/OpenCloudOS-Kernel.git
11
0
Fork 0
Code Issues Projects Releases Wiki Activity
OpenCloudOS-Kernel/arch/sparc/mm/ultra.S

1107 lines
25 KiB
ArmAsm
Raw Normal View History

sparc64: remove CVS keywords This patch removes the CVS keywords that weren't updated for a long time from comments. Signed-off-by: Adrian Bunk <bunk@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-05-20 07:52:27 +08:00
/*
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
* ultra.S: Don't expand these all over the place...
*
sparc64: Add global register dumping facility. When a cpu really is stuck in the kernel, it can be often impossible to figure out which cpu is stuck where. The worst case is when the stuck cpu has interrupts disabled. Therefore, implement a global cpu state capture that uses SMP message interrupts which are not disabled by the normal IRQ enable/disable APIs of the kernel. As long as we can get a sysrq 'y' to the kernel, we can get a dump. Even if the console interrupt cpu is wedged, we can trigger it from userspace using /proc/sysrq-trigger The output is made compact so that this facility is more useful on high cpu count systems, which is where this facility will likely find itself the most useful :) Signed-off-by: David S. Miller <davem@davemloft.net>
2008-05-20 14:46:00 +08:00
* Copyright (C) 1997, 2000, 2008 David S. Miller (davem@davemloft.net)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
#include <asm/asi.h>
#include <asm/pgtable.h>
#include <asm/page.h>
#include <asm/spitfire.h>
#include <asm/mmu_context.h>
[SPARC64]: Preserve nucleus ctx page size during TLB flushes. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-08-31 11:21:34 +08:00
#include <asm/mmu.h>
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
#include <asm/pil.h>
#include <asm/head.h>
#include <asm/thread_info.h>
#include <asm/cacheflush.h>
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
#include <asm/hypervisor.h>
sparc64: Add global register dumping facility. When a cpu really is stuck in the kernel, it can be often impossible to figure out which cpu is stuck where. The worst case is when the stuck cpu has interrupts disabled. Therefore, implement a global cpu state capture that uses SMP message interrupts which are not disabled by the normal IRQ enable/disable APIs of the kernel. As long as we can get a sysrq 'y' to the kernel, we can get a dump. Even if the console interrupt cpu is wedged, we can trigger it from userspace using /proc/sysrq-trigger The output is made compact so that this facility is more useful on high cpu count systems, which is where this facility will likely find itself the most useful :) Signed-off-by: David S. Miller <davem@davemloft.net>
2008-05-20 14:46:00 +08:00
#include <asm/cpudata.h>
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* Basically, most of the Spitfire vs. Cheetah madness
* has to do with the fact that Cheetah does not support
* IMMU flushes out of the secondary context. Someone needs
* to throw a south lake birthday party for the folks
* in Microelectronics who refused to fix this shit.
*/
/* This file is meant to be read efficiently by the CPU, not humans.
* Staraj sie tego nikomu nie pierdolnac...
*/
.text
.align 32
.globl __flush_tlb_mm
sparc64: Fix illegal relative branches in hypervisor patched TLB code. When we copy code over to patch another piece of code, we can only use PC-relative branches that target code within that piece of code. Such PC-relative branches cannot be made to external symbols because the patch moves the location of the code and thus modifies the relative address of external symbols. Use an absolute jmpl to fix this problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-26 07:23:26 +08:00
__flush_tlb_mm: /* 19 insns */
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
/* %o0=(ctx & TAG_CONTEXT_BITS), %o1=SECONDARY_CONTEXT */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
ldxa [%o1] ASI_DMMU, %g2
cmp %g2, %o0
bne,pn %icc, __spitfire_flush_tlb_mm_slow
mov 0x50, %g3
stxa %g0, [%g3] ASI_DMMU_DEMAP
stxa %g0, [%g3] ASI_IMMU_DEMAP
[SPARC64]: Fix bogus flush instruction usage. Some of the trap code was still assuming that alternate global %g6 was hard coded with current_thread_info(). Let's just consistently flush at KERNBASE when we need a pipeline synchronization. That's locked into the TLB and will always work. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-01 10:33:00 +08:00
sethi %hi(KERNBASE), %g3
flush %g3
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
retl
[SPARC64]: Fix bogus flush instruction usage. Some of the trap code was still assuming that alternate global %g6 was hard coded with current_thread_info(). Let's just consistently flush at KERNBASE when we need a pipeline synchronization. That's locked into the TLB and will always work. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-01 10:33:00 +08:00
nop
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
nop
nop
nop
nop
nop
nop
nop
[SPARC64]: Preserve nucleus ctx page size during TLB flushes. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-08-31 11:21:34 +08:00
nop
nop
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
sparc64: Fix race in TLB batch processing. As reported by Dave Kleikamp, when we emit cross calls to do batched TLB flush processing we have a race because we do not synchronize on the sibling cpus completing the cross call. So meanwhile the TLB batch can be reset (tb->tlb_nr set to zero, etc.) and either flushes are missed or flushes will flush the wrong addresses. Fix this by using generic infrastructure to synchonize on the completion of the cross call. This first required getting the flush_tlb_pending() call out from switch_to() which operates with locks held and interrupts disabled. The problem is that smp_call_function_many() cannot be invoked with IRQs disabled and this is explicitly checked for with WARN_ON_ONCE(). We get the batch processing outside of locked IRQ disabled sections by using some ideas from the powerpc port. Namely, we only batch inside of arch_{enter,leave}_lazy_mmu_mode() calls. If we're not in such a region, we flush TLBs synchronously. 1) Get rid of xcall_flush_tlb_pending and per-cpu type implementations. 2) Do TLB batch cross calls instead via: smp_call_function_many() tlb_pending_func() __flush_tlb_pending() 3) Batch only in lazy mmu sequences: a) Add 'active' member to struct tlb_batch b) Define __HAVE_ARCH_ENTER_LAZY_MMU_MODE c) Set 'active' in arch_enter_lazy_mmu_mode() d) Run batch and clear 'active' in arch_leave_lazy_mmu_mode() e) Check 'active' in tlb_batch_add_one() and do a synchronous flush if it's clear. 4) Add infrastructure for synchronous TLB page flushes. a) Implement __flush_tlb_page and per-cpu variants, patch as needed. b) Likewise for xcall_flush_tlb_page. c) Implement smp_flush_tlb_page() to invoke the cross-call. d) Wire up global_flush_tlb_page() to the right routine based upon CONFIG_SMP 5) It turns out that singleton batches are very common, 2 out of every 3 batch flushes have only a single entry in them. The batch flush waiting is very expensive, both because of the poll on sibling cpu completeion, as well as because passing the tlb batch pointer to the sibling cpus invokes a shared memory dereference. Therefore, in flush_tlb_pending(), if there is only one entry in the batch perform a completely asynchronous global_flush_tlb_page() instead. Reported-by: Dave Kleikamp <dave.kleikamp@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Dave Kleikamp <dave.kleikamp@oracle.com>
2013-04-20 05:26:26 +08:00
.align 32
.globl __flush_tlb_page
__flush_tlb_page: /* 22 insns */
/* %o0 = context, %o1 = vaddr */
rdpr %pstate, %g7
andn %g7, PSTATE_IE, %g2
wrpr %g2, %pstate
mov SECONDARY_CONTEXT, %o4
ldxa [%o4] ASI_DMMU, %g2
stxa %o0, [%o4] ASI_DMMU
andcc %o1, 1, %g0
andn %o1, 1, %o3
be,pn %icc, 1f
or %o3, 0x10, %o3
stxa %g0, [%o3] ASI_IMMU_DEMAP
1: stxa %g0, [%o3] ASI_DMMU_DEMAP
membar #Sync
stxa %g2, [%o4] ASI_DMMU
sethi %hi(KERNBASE), %o4
flush %o4
retl
wrpr %g7, 0x0, %pstate
nop
nop
nop
nop
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
.align 32
.globl __flush_tlb_pending
sparc64: Fix illegal relative branches in hypervisor patched TLB code. When we copy code over to patch another piece of code, we can only use PC-relative branches that target code within that piece of code. Such PC-relative branches cannot be made to external symbols because the patch moves the location of the code and thus modifies the relative address of external symbols. Use an absolute jmpl to fix this problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-26 07:23:26 +08:00
__flush_tlb_pending: /* 27 insns */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* %o0 = context, %o1 = nr, %o2 = vaddrs[] */
rdpr %pstate, %g7
sllx %o1, 3, %o1
andn %g7, PSTATE_IE, %g2
wrpr %g2, %pstate
mov SECONDARY_CONTEXT, %o4
ldxa [%o4] ASI_DMMU, %g2
stxa %o0, [%o4] ASI_DMMU
1: sub %o1, (1 << 3), %o1
ldx [%o2 + %o1], %o3
andcc %o3, 1, %g0
andn %o3, 1, %o3
be,pn %icc, 2f
or %o3, 0x10, %o3
stxa %g0, [%o3] ASI_IMMU_DEMAP
2: stxa %g0, [%o3] ASI_DMMU_DEMAP
membar #Sync
brnz,pt %o1, 1b
nop
stxa %g2, [%o4] ASI_DMMU
[SPARC64]: Fix bogus flush instruction usage. Some of the trap code was still assuming that alternate global %g6 was hard coded with current_thread_info(). Let's just consistently flush at KERNBASE when we need a pipeline synchronization. That's locked into the TLB and will always work. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-01 10:33:00 +08:00
sethi %hi(KERNBASE), %o4
flush %o4
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
retl
wrpr %g7, 0x0, %pstate
[SPARC64]: Fix UltraSPARC-III fallout from membar changes. The membar changes made the size of __cheetah_flush_tlb_pending grow by one instruction, but the boot-time code patching was not updated to match. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-07-06 10:45:24 +08:00
nop
[SPARC64]: Preserve nucleus ctx page size during TLB flushes. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-08-31 11:21:34 +08:00
nop
nop
nop
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
.align 32
.globl __flush_tlb_kernel_range
sparc64: Handle extremely large kernel TLB range flushes more gracefully. When the vmalloc area gets fragmented, and because the firmware mapping area sits between where modules live and the vmalloc area, we can sometimes receive requests for enormous kernel TLB range flushes. When this happens the cpu just spins flushing billions of pages and this triggers the NMI watchdog and other problems. We took care of this on the TSB side by doing a linear scan of the table once we pass a certain threshold. Do something similar for the TLB flush, however we are limited by the TLB flush facilities provided by the different chip variants. First of all we use an (mostly arbitrary) cut-off of 256K which is about 32 pages. This can be tuned in the future. The huge range code path for each chip works as follows: 1) On spitfire we flush all non-locked TLB entries using diagnostic acceses. 2) On cheetah we use the "flush all" TLB flush. 3) On sun4v/hypervisor we do a TLB context flush on context 0, which unlike previous chips does not remove "permanent" or locked entries. We could probably do something better on spitfire, such as limiting the flush to kernel TLB entries or even doing range comparisons. However that probably isn't worth it since those chips are old and the TLB only had 64 entries. Reported-by: James Clarke <jrtc27@jrtc27.com> Tested-by: James Clarke <jrtc27@jrtc27.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-28 00:04:54 +08:00
__flush_tlb_kernel_range: /* 31 insns */
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
/* %o0=start, %o1=end */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
cmp %o0, %o1
be,pn %xcc, 2f
sparc64: Handle extremely large kernel TLB range flushes more gracefully. When the vmalloc area gets fragmented, and because the firmware mapping area sits between where modules live and the vmalloc area, we can sometimes receive requests for enormous kernel TLB range flushes. When this happens the cpu just spins flushing billions of pages and this triggers the NMI watchdog and other problems. We took care of this on the TSB side by doing a linear scan of the table once we pass a certain threshold. Do something similar for the TLB flush, however we are limited by the TLB flush facilities provided by the different chip variants. First of all we use an (mostly arbitrary) cut-off of 256K which is about 32 pages. This can be tuned in the future. The huge range code path for each chip works as follows: 1) On spitfire we flush all non-locked TLB entries using diagnostic acceses. 2) On cheetah we use the "flush all" TLB flush. 3) On sun4v/hypervisor we do a TLB context flush on context 0, which unlike previous chips does not remove "permanent" or locked entries. We could probably do something better on spitfire, such as limiting the flush to kernel TLB entries or even doing range comparisons. However that probably isn't worth it since those chips are old and the TLB only had 64 entries. Reported-by: James Clarke <jrtc27@jrtc27.com> Tested-by: James Clarke <jrtc27@jrtc27.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-28 00:04:54 +08:00
sub %o1, %o0, %o3
srlx %o3, 18, %o4
brnz,pn %o4, __spitfire_flush_tlb_kernel_range_slow
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
sethi %hi(PAGE_SIZE), %o4
sub %o3, %o4, %o3
or %o0, 0x20, %o0 ! Nucleus
1: stxa %g0, [%o0 + %o3] ASI_DMMU_DEMAP
stxa %g0, [%o0 + %o3] ASI_IMMU_DEMAP
membar #Sync
brnz,pt %o3, 1b
sub %o3, %o4, %o3
[SPARC64]: Fix bogus flush instruction usage. Some of the trap code was still assuming that alternate global %g6 was hard coded with current_thread_info(). Let's just consistently flush at KERNBASE when we need a pipeline synchronization. That's locked into the TLB and will always work. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-01 10:33:00 +08:00
2: sethi %hi(KERNBASE), %o3
flush %o3
retl
nop
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
nop
sparc64: Fix illegal relative branches in hypervisor patched TLB code. When we copy code over to patch another piece of code, we can only use PC-relative branches that target code within that piece of code. Such PC-relative branches cannot be made to external symbols because the patch moves the location of the code and thus modifies the relative address of external symbols. Use an absolute jmpl to fix this problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-26 07:23:26 +08:00
nop
nop
nop
sparc64: Handle extremely large kernel TLB range flushes more gracefully. When the vmalloc area gets fragmented, and because the firmware mapping area sits between where modules live and the vmalloc area, we can sometimes receive requests for enormous kernel TLB range flushes. When this happens the cpu just spins flushing billions of pages and this triggers the NMI watchdog and other problems. We took care of this on the TSB side by doing a linear scan of the table once we pass a certain threshold. Do something similar for the TLB flush, however we are limited by the TLB flush facilities provided by the different chip variants. First of all we use an (mostly arbitrary) cut-off of 256K which is about 32 pages. This can be tuned in the future. The huge range code path for each chip works as follows: 1) On spitfire we flush all non-locked TLB entries using diagnostic acceses. 2) On cheetah we use the "flush all" TLB flush. 3) On sun4v/hypervisor we do a TLB context flush on context 0, which unlike previous chips does not remove "permanent" or locked entries. We could probably do something better on spitfire, such as limiting the flush to kernel TLB entries or even doing range comparisons. However that probably isn't worth it since those chips are old and the TLB only had 64 entries. Reported-by: James Clarke <jrtc27@jrtc27.com> Tested-by: James Clarke <jrtc27@jrtc27.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-28 00:04:54 +08:00
nop
nop
nop
nop
nop
nop
nop
nop
nop
nop
__spitfire_flush_tlb_kernel_range_slow:
mov 63 * 8, %o4
1: ldxa [%o4] ASI_ITLB_DATA_ACCESS, %o3
andcc %o3, 0x40, %g0 /* _PAGE_L_4U */
bne,pn %xcc, 2f
mov TLB_TAG_ACCESS, %o3
stxa %g0, [%o3] ASI_IMMU
stxa %g0, [%o4] ASI_ITLB_DATA_ACCESS
membar #Sync
2: ldxa [%o4] ASI_DTLB_DATA_ACCESS, %o3
andcc %o3, 0x40, %g0
bne,pn %xcc, 2f
mov TLB_TAG_ACCESS, %o3
stxa %g0, [%o3] ASI_DMMU
stxa %g0, [%o4] ASI_DTLB_DATA_ACCESS
membar #Sync
2: sub %o4, 8, %o4
brgez,pt %o4, 1b
nop
retl
nop
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
__spitfire_flush_tlb_mm_slow:
rdpr %pstate, %g1
wrpr %g1, PSTATE_IE, %pstate
stxa %o0, [%o1] ASI_DMMU
stxa %g0, [%g3] ASI_DMMU_DEMAP
stxa %g0, [%g3] ASI_IMMU_DEMAP
flush %g6
stxa %g2, [%o1] ASI_DMMU
[SPARC64]: Fix bogus flush instruction usage. Some of the trap code was still assuming that alternate global %g6 was hard coded with current_thread_info(). Let's just consistently flush at KERNBASE when we need a pipeline synchronization. That's locked into the TLB and will always work. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-01 10:33:00 +08:00
sethi %hi(KERNBASE), %o1
flush %o1
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
retl
wrpr %g1, 0, %pstate
/*
* The following code flushes one page_size worth.
*/
[PATCH] kprobes-prevent-possible-race-conditions-sparc64-changes fix This patch adds flags "ax" to .kprobe.text section. Signed-off-by: Prasanna S Panchamukhi <prasanna@in.ibm.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-09-07 06:19:31 +08:00
.section .kprobes.text, "ax"
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
.align 32
.globl __flush_icache_page
__flush_icache_page: /* %o0 = phys_page */
srlx %o0, PAGE_SHIFT, %o0
sparc64: Make PAGE_OFFSET variable. Choose PAGE_OFFSET dynamically based upon cpu type. Original UltraSPARC-I (spitfire) chips only supported a 44-bit virtual address space. Newer chips (T4 and later) support 52-bit virtual addresses and up to 47-bits of physical memory space. Therefore we have to adjust PAGE_SIZE dynamically based upon the capabilities of the chip. Note that this change alone does not allow us to support > 43-bit physical memory, to do that we need to re-arrange our page table support. The current encodings of the pmd_t and pgd_t pointers restricts us to "32 + 11" == 43 bits. This change can waste quite a bit of memory for the various tables. In particular, a future change should work to size and allocate kern_linear_bitmap[] and sparc64_valid_addr_bitmap[] dynamically. This isn't easy as we really cannot take a TLB miss when accessing kern_linear_bitmap[]. We'd have to lock it into the TLB or similar. Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Bob Picco <bob.picco@oracle.com>
2013-09-21 12:50:41 +08:00
sethi %hi(PAGE_OFFSET), %g1
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
sllx %o0, PAGE_SHIFT, %o0
sethi %hi(PAGE_SIZE), %g2
sparc64: Make PAGE_OFFSET variable. Choose PAGE_OFFSET dynamically based upon cpu type. Original UltraSPARC-I (spitfire) chips only supported a 44-bit virtual address space. Newer chips (T4 and later) support 52-bit virtual addresses and up to 47-bits of physical memory space. Therefore we have to adjust PAGE_SIZE dynamically based upon the capabilities of the chip. Note that this change alone does not allow us to support > 43-bit physical memory, to do that we need to re-arrange our page table support. The current encodings of the pmd_t and pgd_t pointers restricts us to "32 + 11" == 43 bits. This change can waste quite a bit of memory for the various tables. In particular, a future change should work to size and allocate kern_linear_bitmap[] and sparc64_valid_addr_bitmap[] dynamically. This isn't easy as we really cannot take a TLB miss when accessing kern_linear_bitmap[]. We'd have to lock it into the TLB or similar. Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Bob Picco <bob.picco@oracle.com>
2013-09-21 12:50:41 +08:00
ldx [%g1 + %lo(PAGE_OFFSET)], %g1
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
add %o0, %g1, %o0
1: subcc %g2, 32, %g2
bne,pt %icc, 1b
flush %o0 + %g2
retl
nop
#ifdef DCACHE_ALIASING_POSSIBLE
#if (PAGE_SHIFT != 13)
#error only page shift of 13 is supported by dcache flush
#endif
#define DTAG_MASK 0x3
[SPARC64]: Simplify Spitfire D-cache page flush. It tries to batch up the tag loads and comparisons, and then the stores. And this is just complicated instead of efficient. Also, make the symbol of the Cheetah version more grepable. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-09-27 07:06:03 +08:00
/* This routine is Spitfire specific so the hardcoded
* D-cache size and line-size are OK.
*/
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
.align 64
.globl __flush_dcache_page
__flush_dcache_page: /* %o0=kaddr, %o1=flush_icache */
sparc64: Make PAGE_OFFSET variable. Choose PAGE_OFFSET dynamically based upon cpu type. Original UltraSPARC-I (spitfire) chips only supported a 44-bit virtual address space. Newer chips (T4 and later) support 52-bit virtual addresses and up to 47-bits of physical memory space. Therefore we have to adjust PAGE_SIZE dynamically based upon the capabilities of the chip. Note that this change alone does not allow us to support > 43-bit physical memory, to do that we need to re-arrange our page table support. The current encodings of the pmd_t and pgd_t pointers restricts us to "32 + 11" == 43 bits. This change can waste quite a bit of memory for the various tables. In particular, a future change should work to size and allocate kern_linear_bitmap[] and sparc64_valid_addr_bitmap[] dynamically. This isn't easy as we really cannot take a TLB miss when accessing kern_linear_bitmap[]. We'd have to lock it into the TLB or similar. Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Bob Picco <bob.picco@oracle.com>
2013-09-21 12:50:41 +08:00
sethi %hi(PAGE_OFFSET), %g1
ldx [%g1 + %lo(PAGE_OFFSET)], %g1
[SPARC64]: Simplify Spitfire D-cache page flush. It tries to batch up the tag loads and comparisons, and then the stores. And this is just complicated instead of efficient. Also, make the symbol of the Cheetah version more grepable. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-09-27 07:06:03 +08:00
sub %o0, %g1, %o0 ! physical address
srlx %o0, 11, %o0 ! make D-cache TAG
sethi %hi(1 << 14), %o2 ! D-cache size
sub %o2, (1 << 5), %o2 ! D-cache line size
1: ldxa [%o2] ASI_DCACHE_TAG, %o3 ! load D-cache TAG
andcc %o3, DTAG_MASK, %g0 ! Valid?
be,pn %xcc, 2f ! Nope, branch
andn %o3, DTAG_MASK, %o3 ! Clear valid bits
cmp %o3, %o0 ! TAG match?
bne,pt %xcc, 2f ! Nope, branch
nop
stxa %g0, [%o2] ASI_DCACHE_TAG ! Invalidate TAG
membar #Sync
2: brnz,pt %o2, 1b
sub %o2, (1 << 5), %o2 ! D-cache line size
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* The I-cache does not snoop local stores so we
* better flush that too when necessary.
*/
brnz,pt %o1, __flush_icache_page
sllx %o0, 11, %o0
retl
nop
#endif /* DCACHE_ALIASING_POSSIBLE */
[SPARC64]: Simplify Spitfire D-cache page flush. It tries to batch up the tag loads and comparisons, and then the stores. And this is just complicated instead of efficient. Also, make the symbol of the Cheetah version more grepable. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-09-27 07:06:03 +08:00
.previous
[SPARC64]: Preserve nucleus ctx page size during TLB flushes. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-08-31 11:21:34 +08:00
/* Cheetah specific versions, patched at boot time. */
[SPARC64]: Fix bogus flush instruction usage. Some of the trap code was still assuming that alternate global %g6 was hard coded with current_thread_info(). Let's just consistently flush at KERNBASE when we need a pipeline synchronization. That's locked into the TLB and will always work. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-01 10:33:00 +08:00
__cheetah_flush_tlb_mm: /* 19 insns */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
rdpr %pstate, %g7
andn %g7, PSTATE_IE, %g2
wrpr %g2, 0x0, %pstate
wrpr %g0, 1, %tl
mov PRIMARY_CONTEXT, %o2
mov 0x40, %g3
ldxa [%o2] ASI_DMMU, %g2
[SPARC64]: Preserve nucleus ctx page size during TLB flushes. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-08-31 11:21:34 +08:00
srlx %g2, CTX_PGSZ1_NUC_SHIFT, %o1
sllx %o1, CTX_PGSZ1_NUC_SHIFT, %o1
or %o0, %o1, %o0 /* Preserve nucleus page size fields */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
stxa %o0, [%o2] ASI_DMMU
stxa %g0, [%g3] ASI_DMMU_DEMAP
stxa %g0, [%g3] ASI_IMMU_DEMAP
stxa %g2, [%o2] ASI_DMMU
[SPARC64]: Fix bogus flush instruction usage. Some of the trap code was still assuming that alternate global %g6 was hard coded with current_thread_info(). Let's just consistently flush at KERNBASE when we need a pipeline synchronization. That's locked into the TLB and will always work. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-01 10:33:00 +08:00
sethi %hi(KERNBASE), %o2
flush %o2
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
wrpr %g0, 0, %tl
retl
wrpr %g7, 0x0, %pstate
sparc64: Fix race in TLB batch processing. As reported by Dave Kleikamp, when we emit cross calls to do batched TLB flush processing we have a race because we do not synchronize on the sibling cpus completing the cross call. So meanwhile the TLB batch can be reset (tb->tlb_nr set to zero, etc.) and either flushes are missed or flushes will flush the wrong addresses. Fix this by using generic infrastructure to synchonize on the completion of the cross call. This first required getting the flush_tlb_pending() call out from switch_to() which operates with locks held and interrupts disabled. The problem is that smp_call_function_many() cannot be invoked with IRQs disabled and this is explicitly checked for with WARN_ON_ONCE(). We get the batch processing outside of locked IRQ disabled sections by using some ideas from the powerpc port. Namely, we only batch inside of arch_{enter,leave}_lazy_mmu_mode() calls. If we're not in such a region, we flush TLBs synchronously. 1) Get rid of xcall_flush_tlb_pending and per-cpu type implementations. 2) Do TLB batch cross calls instead via: smp_call_function_many() tlb_pending_func() __flush_tlb_pending() 3) Batch only in lazy mmu sequences: a) Add 'active' member to struct tlb_batch b) Define __HAVE_ARCH_ENTER_LAZY_MMU_MODE c) Set 'active' in arch_enter_lazy_mmu_mode() d) Run batch and clear 'active' in arch_leave_lazy_mmu_mode() e) Check 'active' in tlb_batch_add_one() and do a synchronous flush if it's clear. 4) Add infrastructure for synchronous TLB page flushes. a) Implement __flush_tlb_page and per-cpu variants, patch as needed. b) Likewise for xcall_flush_tlb_page. c) Implement smp_flush_tlb_page() to invoke the cross-call. d) Wire up global_flush_tlb_page() to the right routine based upon CONFIG_SMP 5) It turns out that singleton batches are very common, 2 out of every 3 batch flushes have only a single entry in them. The batch flush waiting is very expensive, both because of the poll on sibling cpu completeion, as well as because passing the tlb batch pointer to the sibling cpus invokes a shared memory dereference. Therefore, in flush_tlb_pending(), if there is only one entry in the batch perform a completely asynchronous global_flush_tlb_page() instead. Reported-by: Dave Kleikamp <dave.kleikamp@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Dave Kleikamp <dave.kleikamp@oracle.com>
2013-04-20 05:26:26 +08:00
__cheetah_flush_tlb_page: /* 22 insns */
/* %o0 = context, %o1 = vaddr */
rdpr %pstate, %g7
andn %g7, PSTATE_IE, %g2
wrpr %g2, 0x0, %pstate
wrpr %g0, 1, %tl
mov PRIMARY_CONTEXT, %o4
ldxa [%o4] ASI_DMMU, %g2
srlx %g2, CTX_PGSZ1_NUC_SHIFT, %o3
sllx %o3, CTX_PGSZ1_NUC_SHIFT, %o3
or %o0, %o3, %o0 /* Preserve nucleus page size fields */
stxa %o0, [%o4] ASI_DMMU
andcc %o1, 1, %g0
be,pn %icc, 1f
andn %o1, 1, %o3
stxa %g0, [%o3] ASI_IMMU_DEMAP
1: stxa %g0, [%o3] ASI_DMMU_DEMAP
membar #Sync
stxa %g2, [%o4] ASI_DMMU
sethi %hi(KERNBASE), %o4
flush %o4
wrpr %g0, 0, %tl
retl
wrpr %g7, 0x0, %pstate
[SPARC64]: Fix bogus flush instruction usage. Some of the trap code was still assuming that alternate global %g6 was hard coded with current_thread_info(). Let's just consistently flush at KERNBASE when we need a pipeline synchronization. That's locked into the TLB and will always work. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-01 10:33:00 +08:00
__cheetah_flush_tlb_pending: /* 27 insns */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* %o0 = context, %o1 = nr, %o2 = vaddrs[] */
rdpr %pstate, %g7
sllx %o1, 3, %o1
andn %g7, PSTATE_IE, %g2
wrpr %g2, 0x0, %pstate
wrpr %g0, 1, %tl
mov PRIMARY_CONTEXT, %o4
ldxa [%o4] ASI_DMMU, %g2
[SPARC64]: Preserve nucleus ctx page size during TLB flushes. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-08-31 11:21:34 +08:00
srlx %g2, CTX_PGSZ1_NUC_SHIFT, %o3
sllx %o3, CTX_PGSZ1_NUC_SHIFT, %o3
or %o0, %o3, %o0 /* Preserve nucleus page size fields */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
stxa %o0, [%o4] ASI_DMMU
1: sub %o1, (1 << 3), %o1
ldx [%o2 + %o1], %o3
andcc %o3, 1, %g0
be,pn %icc, 2f
andn %o3, 1, %o3
stxa %g0, [%o3] ASI_IMMU_DEMAP
2: stxa %g0, [%o3] ASI_DMMU_DEMAP
[SPARC64]: Avoid membar instructions in delay slots. In particular, avoid membar instructions in the delay slot of a jmpl instruction. UltraSPARC-I, II, IIi, and IIe have a bug, documented in the UltraSPARC-IIi User's Manual, Appendix K, Erratum 51 The long and short of it is that if the IMU unit misses on a branch or jmpl, and there is a store buffer synchronizing membar in the delay slot, the chip can stop fetching instructions. If interrupts are enabled or some other trap is enabled, the chip will unwedge itself, but performance will suffer. We already had a workaround for this bug in a few spots, but it's better to have the entire tree sanitized for this rule. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-06-28 06:42:04 +08:00
membar #Sync
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
brnz,pt %o1, 1b
[SPARC64]: Avoid membar instructions in delay slots. In particular, avoid membar instructions in the delay slot of a jmpl instruction. UltraSPARC-I, II, IIi, and IIe have a bug, documented in the UltraSPARC-IIi User's Manual, Appendix K, Erratum 51 The long and short of it is that if the IMU unit misses on a branch or jmpl, and there is a store buffer synchronizing membar in the delay slot, the chip can stop fetching instructions. If interrupts are enabled or some other trap is enabled, the chip will unwedge itself, but performance will suffer. We already had a workaround for this bug in a few spots, but it's better to have the entire tree sanitized for this rule. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-06-28 06:42:04 +08:00
nop
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
stxa %g2, [%o4] ASI_DMMU
[SPARC64]: Fix bogus flush instruction usage. Some of the trap code was still assuming that alternate global %g6 was hard coded with current_thread_info(). Let's just consistently flush at KERNBASE when we need a pipeline synchronization. That's locked into the TLB and will always work. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-01 10:33:00 +08:00
sethi %hi(KERNBASE), %o4
flush %o4
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
wrpr %g0, 0, %tl
retl
wrpr %g7, 0x0, %pstate
sparc64: Handle extremely large kernel TLB range flushes more gracefully. When the vmalloc area gets fragmented, and because the firmware mapping area sits between where modules live and the vmalloc area, we can sometimes receive requests for enormous kernel TLB range flushes. When this happens the cpu just spins flushing billions of pages and this triggers the NMI watchdog and other problems. We took care of this on the TSB side by doing a linear scan of the table once we pass a certain threshold. Do something similar for the TLB flush, however we are limited by the TLB flush facilities provided by the different chip variants. First of all we use an (mostly arbitrary) cut-off of 256K which is about 32 pages. This can be tuned in the future. The huge range code path for each chip works as follows: 1) On spitfire we flush all non-locked TLB entries using diagnostic acceses. 2) On cheetah we use the "flush all" TLB flush. 3) On sun4v/hypervisor we do a TLB context flush on context 0, which unlike previous chips does not remove "permanent" or locked entries. We could probably do something better on spitfire, such as limiting the flush to kernel TLB entries or even doing range comparisons. However that probably isn't worth it since those chips are old and the TLB only had 64 entries. Reported-by: James Clarke <jrtc27@jrtc27.com> Tested-by: James Clarke <jrtc27@jrtc27.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-28 00:04:54 +08:00
__cheetah_flush_tlb_kernel_range: /* 31 insns */
/* %o0=start, %o1=end */
cmp %o0, %o1
be,pn %xcc, 2f
sub %o1, %o0, %o3
srlx %o3, 18, %o4
brnz,pn %o4, 3f
sethi %hi(PAGE_SIZE), %o4
sub %o3, %o4, %o3
or %o0, 0x20, %o0 ! Nucleus
1: stxa %g0, [%o0 + %o3] ASI_DMMU_DEMAP
stxa %g0, [%o0 + %o3] ASI_IMMU_DEMAP
membar #Sync
brnz,pt %o3, 1b
sub %o3, %o4, %o3
2: sethi %hi(KERNBASE), %o3
flush %o3
retl
nop
3: mov 0x80, %o4
stxa %g0, [%o4] ASI_DMMU_DEMAP
membar #Sync
stxa %g0, [%o4] ASI_IMMU_DEMAP
membar #Sync
retl
nop
nop
nop
nop
nop
nop
nop
nop
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
#ifdef DCACHE_ALIASING_POSSIBLE
[SPARC64]: Simplify Spitfire D-cache page flush. It tries to batch up the tag loads and comparisons, and then the stores. And this is just complicated instead of efficient. Also, make the symbol of the Cheetah version more grepable. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-09-27 07:06:03 +08:00
__cheetah_flush_dcache_page: /* 11 insns */
sparc64: Make PAGE_OFFSET variable. Choose PAGE_OFFSET dynamically based upon cpu type. Original UltraSPARC-I (spitfire) chips only supported a 44-bit virtual address space. Newer chips (T4 and later) support 52-bit virtual addresses and up to 47-bits of physical memory space. Therefore we have to adjust PAGE_SIZE dynamically based upon the capabilities of the chip. Note that this change alone does not allow us to support > 43-bit physical memory, to do that we need to re-arrange our page table support. The current encodings of the pmd_t and pgd_t pointers restricts us to "32 + 11" == 43 bits. This change can waste quite a bit of memory for the various tables. In particular, a future change should work to size and allocate kern_linear_bitmap[] and sparc64_valid_addr_bitmap[] dynamically. This isn't easy as we really cannot take a TLB miss when accessing kern_linear_bitmap[]. We'd have to lock it into the TLB or similar. Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Bob Picco <bob.picco@oracle.com>
2013-09-21 12:50:41 +08:00
sethi %hi(PAGE_OFFSET), %g1
ldx [%g1 + %lo(PAGE_OFFSET)], %g1
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
sub %o0, %g1, %o0
sethi %hi(PAGE_SIZE), %o4
1: subcc %o4, (1 << 5), %o4
stxa %g0, [%o0 + %o4] ASI_DCACHE_INVALIDATE
membar #Sync
bne,pt %icc, 1b
nop
retl /* I-cache flush never needed on Cheetah, see callers. */
nop
#endif /* DCACHE_ALIASING_POSSIBLE */
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
/* Hypervisor specific versions, patched at boot time. */
[SPARC64]: Bulletproof hypervisor TLB flushing. Check TLB flush hypervisor calls for errors and report them. Pass HV_MMU_ALL always for now, we can add back the optimization to avoid the I-TLB flush later. Always explicitly page align the virtual address arguments. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-27 11:31:49 +08:00
__hypervisor_tlb_tl0_error:
save %sp, -192, %sp
mov %i0, %o0
call hypervisor_tlbop_error
mov %i1, %o1
ret
restore
sparc64: Fix illegal relative branches in hypervisor patched TLB code. When we copy code over to patch another piece of code, we can only use PC-relative branches that target code within that piece of code. Such PC-relative branches cannot be made to external symbols because the patch moves the location of the code and thus modifies the relative address of external symbols. Use an absolute jmpl to fix this problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-26 07:23:26 +08:00
__hypervisor_flush_tlb_mm: /* 19 insns */
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
mov %o0, %o2 /* ARG2: mmu context */
mov 0, %o0 /* ARG0: CPU lists unimplemented */
mov 0, %o1 /* ARG1: CPU lists unimplemented */
mov HV_MMU_ALL, %o3 /* ARG3: flags */
mov HV_FAST_MMU_DEMAP_CTX, %o5
ta HV_FAST_TRAP
sparc64: Fix illegal relative branches in hypervisor patched TLB code. When we copy code over to patch another piece of code, we can only use PC-relative branches that target code within that piece of code. Such PC-relative branches cannot be made to external symbols because the patch moves the location of the code and thus modifies the relative address of external symbols. Use an absolute jmpl to fix this problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-26 07:23:26 +08:00
brnz,pn %o0, 1f
[SPARC64]: Bulletproof hypervisor TLB flushing. Check TLB flush hypervisor calls for errors and report them. Pass HV_MMU_ALL always for now, we can add back the optimization to avoid the I-TLB flush later. Always explicitly page align the virtual address arguments. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-27 11:31:49 +08:00
mov HV_FAST_MMU_DEMAP_CTX, %o1
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
retl
nop
sparc64: Fix illegal relative branches in hypervisor patched TLB code. When we copy code over to patch another piece of code, we can only use PC-relative branches that target code within that piece of code. Such PC-relative branches cannot be made to external symbols because the patch moves the location of the code and thus modifies the relative address of external symbols. Use an absolute jmpl to fix this problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-26 07:23:26 +08:00
1: sethi %hi(__hypervisor_tlb_tl0_error), %o5
jmpl %o5 + %lo(__hypervisor_tlb_tl0_error), %g0
nop
nop
nop
nop
nop
nop
nop
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
sparc64: Fix illegal relative branches in hypervisor patched TLB code. When we copy code over to patch another piece of code, we can only use PC-relative branches that target code within that piece of code. Such PC-relative branches cannot be made to external symbols because the patch moves the location of the code and thus modifies the relative address of external symbols. Use an absolute jmpl to fix this problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-26 07:23:26 +08:00
__hypervisor_flush_tlb_page: /* 22 insns */
sparc64: Fix race in TLB batch processing. As reported by Dave Kleikamp, when we emit cross calls to do batched TLB flush processing we have a race because we do not synchronize on the sibling cpus completing the cross call. So meanwhile the TLB batch can be reset (tb->tlb_nr set to zero, etc.) and either flushes are missed or flushes will flush the wrong addresses. Fix this by using generic infrastructure to synchonize on the completion of the cross call. This first required getting the flush_tlb_pending() call out from switch_to() which operates with locks held and interrupts disabled. The problem is that smp_call_function_many() cannot be invoked with IRQs disabled and this is explicitly checked for with WARN_ON_ONCE(). We get the batch processing outside of locked IRQ disabled sections by using some ideas from the powerpc port. Namely, we only batch inside of arch_{enter,leave}_lazy_mmu_mode() calls. If we're not in such a region, we flush TLBs synchronously. 1) Get rid of xcall_flush_tlb_pending and per-cpu type implementations. 2) Do TLB batch cross calls instead via: smp_call_function_many() tlb_pending_func() __flush_tlb_pending() 3) Batch only in lazy mmu sequences: a) Add 'active' member to struct tlb_batch b) Define __HAVE_ARCH_ENTER_LAZY_MMU_MODE c) Set 'active' in arch_enter_lazy_mmu_mode() d) Run batch and clear 'active' in arch_leave_lazy_mmu_mode() e) Check 'active' in tlb_batch_add_one() and do a synchronous flush if it's clear. 4) Add infrastructure for synchronous TLB page flushes. a) Implement __flush_tlb_page and per-cpu variants, patch as needed. b) Likewise for xcall_flush_tlb_page. c) Implement smp_flush_tlb_page() to invoke the cross-call. d) Wire up global_flush_tlb_page() to the right routine based upon CONFIG_SMP 5) It turns out that singleton batches are very common, 2 out of every 3 batch flushes have only a single entry in them. The batch flush waiting is very expensive, both because of the poll on sibling cpu completeion, as well as because passing the tlb batch pointer to the sibling cpus invokes a shared memory dereference. Therefore, in flush_tlb_pending(), if there is only one entry in the batch perform a completely asynchronous global_flush_tlb_page() instead. Reported-by: Dave Kleikamp <dave.kleikamp@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Dave Kleikamp <dave.kleikamp@oracle.com>
2013-04-20 05:26:26 +08:00
/* %o0 = context, %o1 = vaddr */
mov %o0, %g2
mov %o1, %o0 /* ARG0: vaddr + IMMU-bit */
mov %g2, %o1 /* ARG1: mmu context */
mov HV_MMU_ALL, %o2 /* ARG2: flags */
srlx %o0, PAGE_SHIFT, %o0
sllx %o0, PAGE_SHIFT, %o0
ta HV_MMU_UNMAP_ADDR_TRAP
sparc64: Fix illegal relative branches in hypervisor patched TLB code. When we copy code over to patch another piece of code, we can only use PC-relative branches that target code within that piece of code. Such PC-relative branches cannot be made to external symbols because the patch moves the location of the code and thus modifies the relative address of external symbols. Use an absolute jmpl to fix this problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-26 07:23:26 +08:00
brnz,pn %o0, 1f
sparc64: Fix race in TLB batch processing. As reported by Dave Kleikamp, when we emit cross calls to do batched TLB flush processing we have a race because we do not synchronize on the sibling cpus completing the cross call. So meanwhile the TLB batch can be reset (tb->tlb_nr set to zero, etc.) and either flushes are missed or flushes will flush the wrong addresses. Fix this by using generic infrastructure to synchonize on the completion of the cross call. This first required getting the flush_tlb_pending() call out from switch_to() which operates with locks held and interrupts disabled. The problem is that smp_call_function_many() cannot be invoked with IRQs disabled and this is explicitly checked for with WARN_ON_ONCE(). We get the batch processing outside of locked IRQ disabled sections by using some ideas from the powerpc port. Namely, we only batch inside of arch_{enter,leave}_lazy_mmu_mode() calls. If we're not in such a region, we flush TLBs synchronously. 1) Get rid of xcall_flush_tlb_pending and per-cpu type implementations. 2) Do TLB batch cross calls instead via: smp_call_function_many() tlb_pending_func() __flush_tlb_pending() 3) Batch only in lazy mmu sequences: a) Add 'active' member to struct tlb_batch b) Define __HAVE_ARCH_ENTER_LAZY_MMU_MODE c) Set 'active' in arch_enter_lazy_mmu_mode() d) Run batch and clear 'active' in arch_leave_lazy_mmu_mode() e) Check 'active' in tlb_batch_add_one() and do a synchronous flush if it's clear. 4) Add infrastructure for synchronous TLB page flushes. a) Implement __flush_tlb_page and per-cpu variants, patch as needed. b) Likewise for xcall_flush_tlb_page. c) Implement smp_flush_tlb_page() to invoke the cross-call. d) Wire up global_flush_tlb_page() to the right routine based upon CONFIG_SMP 5) It turns out that singleton batches are very common, 2 out of every 3 batch flushes have only a single entry in them. The batch flush waiting is very expensive, both because of the poll on sibling cpu completeion, as well as because passing the tlb batch pointer to the sibling cpus invokes a shared memory dereference. Therefore, in flush_tlb_pending(), if there is only one entry in the batch perform a completely asynchronous global_flush_tlb_page() instead. Reported-by: Dave Kleikamp <dave.kleikamp@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Dave Kleikamp <dave.kleikamp@oracle.com>
2013-04-20 05:26:26 +08:00
mov HV_MMU_UNMAP_ADDR_TRAP, %o1
retl
nop
sparc64: Fix illegal relative branches in hypervisor patched TLB code. When we copy code over to patch another piece of code, we can only use PC-relative branches that target code within that piece of code. Such PC-relative branches cannot be made to external symbols because the patch moves the location of the code and thus modifies the relative address of external symbols. Use an absolute jmpl to fix this problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-26 07:23:26 +08:00
1: sethi %hi(__hypervisor_tlb_tl0_error), %o2
jmpl %o2 + %lo(__hypervisor_tlb_tl0_error), %g0
nop
nop
nop
nop
nop
nop
nop
nop
nop
sparc64: Fix race in TLB batch processing. As reported by Dave Kleikamp, when we emit cross calls to do batched TLB flush processing we have a race because we do not synchronize on the sibling cpus completing the cross call. So meanwhile the TLB batch can be reset (tb->tlb_nr set to zero, etc.) and either flushes are missed or flushes will flush the wrong addresses. Fix this by using generic infrastructure to synchonize on the completion of the cross call. This first required getting the flush_tlb_pending() call out from switch_to() which operates with locks held and interrupts disabled. The problem is that smp_call_function_many() cannot be invoked with IRQs disabled and this is explicitly checked for with WARN_ON_ONCE(). We get the batch processing outside of locked IRQ disabled sections by using some ideas from the powerpc port. Namely, we only batch inside of arch_{enter,leave}_lazy_mmu_mode() calls. If we're not in such a region, we flush TLBs synchronously. 1) Get rid of xcall_flush_tlb_pending and per-cpu type implementations. 2) Do TLB batch cross calls instead via: smp_call_function_many() tlb_pending_func() __flush_tlb_pending() 3) Batch only in lazy mmu sequences: a) Add 'active' member to struct tlb_batch b) Define __HAVE_ARCH_ENTER_LAZY_MMU_MODE c) Set 'active' in arch_enter_lazy_mmu_mode() d) Run batch and clear 'active' in arch_leave_lazy_mmu_mode() e) Check 'active' in tlb_batch_add_one() and do a synchronous flush if it's clear. 4) Add infrastructure for synchronous TLB page flushes. a) Implement __flush_tlb_page and per-cpu variants, patch as needed. b) Likewise for xcall_flush_tlb_page. c) Implement smp_flush_tlb_page() to invoke the cross-call. d) Wire up global_flush_tlb_page() to the right routine based upon CONFIG_SMP 5) It turns out that singleton batches are very common, 2 out of every 3 batch flushes have only a single entry in them. The batch flush waiting is very expensive, both because of the poll on sibling cpu completeion, as well as because passing the tlb batch pointer to the sibling cpus invokes a shared memory dereference. Therefore, in flush_tlb_pending(), if there is only one entry in the batch perform a completely asynchronous global_flush_tlb_page() instead. Reported-by: Dave Kleikamp <dave.kleikamp@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Dave Kleikamp <dave.kleikamp@oracle.com>
2013-04-20 05:26:26 +08:00
sparc64: Fix instruction count in comment for __hypervisor_flush_tlb_pending. Noticed by James Clarke. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-27 01:08:22 +08:00
__hypervisor_flush_tlb_pending: /* 27 insns */
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
/* %o0 = context, %o1 = nr, %o2 = vaddrs[] */
sllx %o1, 3, %g1
mov %o2, %g2
mov %o0, %g3
1: sub %g1, (1 << 3), %g1
ldx [%g2 + %g1], %o0 /* ARG0: vaddr + IMMU-bit */
mov %g3, %o1 /* ARG1: mmu context */
[SPARC64]: Bulletproof hypervisor TLB flushing. Check TLB flush hypervisor calls for errors and report them. Pass HV_MMU_ALL always for now, we can add back the optimization to avoid the I-TLB flush later. Always explicitly page align the virtual address arguments. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-27 11:31:49 +08:00
mov HV_MMU_ALL, %o2 /* ARG2: flags */
srlx %o0, PAGE_SHIFT, %o0
sllx %o0, PAGE_SHIFT, %o0
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
ta HV_MMU_UNMAP_ADDR_TRAP
sparc64: Fix illegal relative branches in hypervisor patched TLB code. When we copy code over to patch another piece of code, we can only use PC-relative branches that target code within that piece of code. Such PC-relative branches cannot be made to external symbols because the patch moves the location of the code and thus modifies the relative address of external symbols. Use an absolute jmpl to fix this problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-26 07:23:26 +08:00
brnz,pn %o0, 1f
[SPARC64]: Bulletproof hypervisor TLB flushing. Check TLB flush hypervisor calls for errors and report them. Pass HV_MMU_ALL always for now, we can add back the optimization to avoid the I-TLB flush later. Always explicitly page align the virtual address arguments. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-27 11:31:49 +08:00
mov HV_MMU_UNMAP_ADDR_TRAP, %o1
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
brnz,pt %g1, 1b
nop
retl
nop
sparc64: Fix illegal relative branches in hypervisor patched TLB code. When we copy code over to patch another piece of code, we can only use PC-relative branches that target code within that piece of code. Such PC-relative branches cannot be made to external symbols because the patch moves the location of the code and thus modifies the relative address of external symbols. Use an absolute jmpl to fix this problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-26 07:23:26 +08:00
1: sethi %hi(__hypervisor_tlb_tl0_error), %o2
jmpl %o2 + %lo(__hypervisor_tlb_tl0_error), %g0
nop
nop
nop
nop
nop
nop
nop
nop
nop
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
sparc64: Handle extremely large kernel TLB range flushes more gracefully. When the vmalloc area gets fragmented, and because the firmware mapping area sits between where modules live and the vmalloc area, we can sometimes receive requests for enormous kernel TLB range flushes. When this happens the cpu just spins flushing billions of pages and this triggers the NMI watchdog and other problems. We took care of this on the TSB side by doing a linear scan of the table once we pass a certain threshold. Do something similar for the TLB flush, however we are limited by the TLB flush facilities provided by the different chip variants. First of all we use an (mostly arbitrary) cut-off of 256K which is about 32 pages. This can be tuned in the future. The huge range code path for each chip works as follows: 1) On spitfire we flush all non-locked TLB entries using diagnostic acceses. 2) On cheetah we use the "flush all" TLB flush. 3) On sun4v/hypervisor we do a TLB context flush on context 0, which unlike previous chips does not remove "permanent" or locked entries. We could probably do something better on spitfire, such as limiting the flush to kernel TLB entries or even doing range comparisons. However that probably isn't worth it since those chips are old and the TLB only had 64 entries. Reported-by: James Clarke <jrtc27@jrtc27.com> Tested-by: James Clarke <jrtc27@jrtc27.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-28 00:04:54 +08:00
__hypervisor_flush_tlb_kernel_range: /* 31 insns */
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
/* %o0=start, %o1=end */
cmp %o0, %o1
be,pn %xcc, 2f
sparc64: Handle extremely large kernel TLB range flushes more gracefully. When the vmalloc area gets fragmented, and because the firmware mapping area sits between where modules live and the vmalloc area, we can sometimes receive requests for enormous kernel TLB range flushes. When this happens the cpu just spins flushing billions of pages and this triggers the NMI watchdog and other problems. We took care of this on the TSB side by doing a linear scan of the table once we pass a certain threshold. Do something similar for the TLB flush, however we are limited by the TLB flush facilities provided by the different chip variants. First of all we use an (mostly arbitrary) cut-off of 256K which is about 32 pages. This can be tuned in the future. The huge range code path for each chip works as follows: 1) On spitfire we flush all non-locked TLB entries using diagnostic acceses. 2) On cheetah we use the "flush all" TLB flush. 3) On sun4v/hypervisor we do a TLB context flush on context 0, which unlike previous chips does not remove "permanent" or locked entries. We could probably do something better on spitfire, such as limiting the flush to kernel TLB entries or even doing range comparisons. However that probably isn't worth it since those chips are old and the TLB only had 64 entries. Reported-by: James Clarke <jrtc27@jrtc27.com> Tested-by: James Clarke <jrtc27@jrtc27.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-28 00:04:54 +08:00
sub %o1, %o0, %g2
srlx %g2, 18, %g3
brnz,pn %g3, 4f
mov %o0, %g1
sethi %hi(PAGE_SIZE), %g3
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
sub %g2, %g3, %g2
1: add %g1, %g2, %o0 /* ARG0: virtual address */
mov 0, %o1 /* ARG1: mmu context */
mov HV_MMU_ALL, %o2 /* ARG2: flags */
ta HV_MMU_UNMAP_ADDR_TRAP
sparc64: Fix illegal relative branches in hypervisor patched TLB code. When we copy code over to patch another piece of code, we can only use PC-relative branches that target code within that piece of code. Such PC-relative branches cannot be made to external symbols because the patch moves the location of the code and thus modifies the relative address of external symbols. Use an absolute jmpl to fix this problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-26 07:23:26 +08:00
brnz,pn %o0, 3f
[SPARC64]: Bulletproof hypervisor TLB flushing. Check TLB flush hypervisor calls for errors and report them. Pass HV_MMU_ALL always for now, we can add back the optimization to avoid the I-TLB flush later. Always explicitly page align the virtual address arguments. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-27 11:31:49 +08:00
mov HV_MMU_UNMAP_ADDR_TRAP, %o1
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
brnz,pt %g2, 1b
sub %g2, %g3, %g2
2: retl
nop
sparc64: Fix illegal relative branches in hypervisor patched TLB code. When we copy code over to patch another piece of code, we can only use PC-relative branches that target code within that piece of code. Such PC-relative branches cannot be made to external symbols because the patch moves the location of the code and thus modifies the relative address of external symbols. Use an absolute jmpl to fix this problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-26 07:23:26 +08:00
3: sethi %hi(__hypervisor_tlb_tl0_error), %o2
jmpl %o2 + %lo(__hypervisor_tlb_tl0_error), %g0
nop
sparc64: Handle extremely large kernel TLB range flushes more gracefully. When the vmalloc area gets fragmented, and because the firmware mapping area sits between where modules live and the vmalloc area, we can sometimes receive requests for enormous kernel TLB range flushes. When this happens the cpu just spins flushing billions of pages and this triggers the NMI watchdog and other problems. We took care of this on the TSB side by doing a linear scan of the table once we pass a certain threshold. Do something similar for the TLB flush, however we are limited by the TLB flush facilities provided by the different chip variants. First of all we use an (mostly arbitrary) cut-off of 256K which is about 32 pages. This can be tuned in the future. The huge range code path for each chip works as follows: 1) On spitfire we flush all non-locked TLB entries using diagnostic acceses. 2) On cheetah we use the "flush all" TLB flush. 3) On sun4v/hypervisor we do a TLB context flush on context 0, which unlike previous chips does not remove "permanent" or locked entries. We could probably do something better on spitfire, such as limiting the flush to kernel TLB entries or even doing range comparisons. However that probably isn't worth it since those chips are old and the TLB only had 64 entries. Reported-by: James Clarke <jrtc27@jrtc27.com> Tested-by: James Clarke <jrtc27@jrtc27.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-28 00:04:54 +08:00
4: mov 0, %o0 /* ARG0: CPU lists unimplemented */
mov 0, %o1 /* ARG1: CPU lists unimplemented */
mov 0, %o2 /* ARG2: mmu context == nucleus */
mov HV_MMU_ALL, %o3 /* ARG3: flags */
mov HV_FAST_MMU_DEMAP_CTX, %o5
ta HV_FAST_TRAP
brnz,pn %o0, 3b
mov HV_FAST_MMU_DEMAP_CTX, %o1
retl
nop
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
#ifdef DCACHE_ALIASING_POSSIBLE
/* XXX Niagara and friends have an 8K cache, so no aliasing is
* XXX possible, but nothing explicit in the Hypervisor API
* XXX guarantees this.
*/
__hypervisor_flush_dcache_page: /* 2 insns */
retl
nop
#endif
tlb_patch_one:
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
1: lduw [%o1], %g1
stw %g1, [%o0]
flush %o0
subcc %o2, 1, %o2
add %o1, 4, %o1
bne,pt %icc, 1b
add %o0, 4, %o0
retl
nop
#ifdef CONFIG_SMP
/* These are all called by the slaves of a cross call, at
* trap level 1, with interrupts fully disabled.
*
* Register usage:
* %g5 mm->context (all tlb flushes)
* %g1 address arg 1 (tlb page and range flushes)
* %g7 address arg 2 (tlb range flush only)
*
[SPARC64]: Elminate all usage of hard-coded trap globals. UltraSPARC has special sets of global registers which are switched to for certain trap types. There is one set for MMU related traps, one set of Interrupt Vector processing, and another set (called the Alternate globals) for all other trap types. For what seems like forever we've hard coded the values in some of these trap registers. Some examples include: 1) Interrupt Vector global %g6 holds current processors interrupt work struct where received interrupts are managed for IRQ handler dispatch. 2) MMU global %g7 holds the base of the page tables of the currently active address space. 3) Alternate global %g6 held the current_thread_info() value. Such hardcoding has resulted in some serious issues in many areas. There are some code sequences where having another register available would help clean up the implementation. Taking traps such as cross-calls from the OBP firmware requires some trick code sequences wherein we have to save away and restore all of the special sets of global registers when we enter/exit OBP. We were also using the IMMU TSB register on SMP to hold the per-cpu area base address, which doesn't work any longer now that we actually use the TSB facility of the cpu. The implementation is pretty straight forward. One tricky bit is getting the current processor ID as that is different on different cpu variants. We use a stub with a fancy calling convention which we patch at boot time. The calling convention is that the stub is branched to and the (PC - 4) to return to is in register %g1. The cpu number is left in %g6. This stub can be invoked by using the __GET_CPUID macro. We use an array of per-cpu trap state to store the current thread and physical address of the current address space's page tables. The TRAP_LOAD_THREAD_REG loads %g6 with the current thread from this table, it uses __GET_CPUID and also clobbers %g1. TRAP_LOAD_IRQ_WORK is used by the interrupt vector processing to load the current processor's IRQ software state into %g6. It also uses __GET_CPUID and clobbers %g1. Finally, TRAP_LOAD_PGD_PHYS loads the physical address base of the current address space's page tables into %g7, it clobbers %g1 and uses __GET_CPUID. Many refinements are possible, as well as some tuning, with this stuff in place. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-27 15:24:22 +08:00
* %g6 scratch 1
* %g2 scratch 2
* %g3 scratch 3
* %g4 scratch 4
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
.align 32
.globl xcall_flush_tlb_mm
sparc64: Fix illegal relative branches in hypervisor patched TLB cross-call code. Just like the non-cross-call TLB flush handlers, the cross-call ones need to avoid doing PC-relative branches outside of their code blocks. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-27 01:20:14 +08:00
xcall_flush_tlb_mm: /* 24 insns */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
mov PRIMARY_CONTEXT, %g2
ldxa [%g2] ASI_DMMU, %g3
[SPARC64]: Preserve nucleus ctx page size during TLB flushes. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-08-31 11:21:34 +08:00
srlx %g3, CTX_PGSZ1_NUC_SHIFT, %g4
sllx %g4, CTX_PGSZ1_NUC_SHIFT, %g4
or %g5, %g4, %g5 /* Preserve nucleus page size fields */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
stxa %g5, [%g2] ASI_DMMU
[SPARC64]: Preserve nucleus ctx page size during TLB flushes. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-08-31 11:21:34 +08:00
mov 0x40, %g4
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
stxa %g0, [%g4] ASI_DMMU_DEMAP
stxa %g0, [%g4] ASI_IMMU_DEMAP
stxa %g3, [%g2] ASI_DMMU
retry
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
nop
nop
nop
nop
nop
nop
nop
[SPARC64]: Bulletproof hypervisor TLB flushing. Check TLB flush hypervisor calls for errors and report them. Pass HV_MMU_ALL always for now, we can add back the optimization to avoid the I-TLB flush later. Always explicitly page align the virtual address arguments. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-27 11:31:49 +08:00
nop
nop
nop
sparc64: Fix illegal relative branches in hypervisor patched TLB cross-call code. Just like the non-cross-call TLB flush handlers, the cross-call ones need to avoid doing PC-relative branches outside of their code blocks. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-27 01:20:14 +08:00
nop
nop
nop
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
sparc64: Fix race in TLB batch processing. As reported by Dave Kleikamp, when we emit cross calls to do batched TLB flush processing we have a race because we do not synchronize on the sibling cpus completing the cross call. So meanwhile the TLB batch can be reset (tb->tlb_nr set to zero, etc.) and either flushes are missed or flushes will flush the wrong addresses. Fix this by using generic infrastructure to synchonize on the completion of the cross call. This first required getting the flush_tlb_pending() call out from switch_to() which operates with locks held and interrupts disabled. The problem is that smp_call_function_many() cannot be invoked with IRQs disabled and this is explicitly checked for with WARN_ON_ONCE(). We get the batch processing outside of locked IRQ disabled sections by using some ideas from the powerpc port. Namely, we only batch inside of arch_{enter,leave}_lazy_mmu_mode() calls. If we're not in such a region, we flush TLBs synchronously. 1) Get rid of xcall_flush_tlb_pending and per-cpu type implementations. 2) Do TLB batch cross calls instead via: smp_call_function_many() tlb_pending_func() __flush_tlb_pending() 3) Batch only in lazy mmu sequences: a) Add 'active' member to struct tlb_batch b) Define __HAVE_ARCH_ENTER_LAZY_MMU_MODE c) Set 'active' in arch_enter_lazy_mmu_mode() d) Run batch and clear 'active' in arch_leave_lazy_mmu_mode() e) Check 'active' in tlb_batch_add_one() and do a synchronous flush if it's clear. 4) Add infrastructure for synchronous TLB page flushes. a) Implement __flush_tlb_page and per-cpu variants, patch as needed. b) Likewise for xcall_flush_tlb_page. c) Implement smp_flush_tlb_page() to invoke the cross-call. d) Wire up global_flush_tlb_page() to the right routine based upon CONFIG_SMP 5) It turns out that singleton batches are very common, 2 out of every 3 batch flushes have only a single entry in them. The batch flush waiting is very expensive, both because of the poll on sibling cpu completeion, as well as because passing the tlb batch pointer to the sibling cpus invokes a shared memory dereference. Therefore, in flush_tlb_pending(), if there is only one entry in the batch perform a completely asynchronous global_flush_tlb_page() instead. Reported-by: Dave Kleikamp <dave.kleikamp@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Dave Kleikamp <dave.kleikamp@oracle.com>
2013-04-20 05:26:26 +08:00
.globl xcall_flush_tlb_page
sparc64: Fix illegal relative branches in hypervisor patched TLB cross-call code. Just like the non-cross-call TLB flush handlers, the cross-call ones need to avoid doing PC-relative branches outside of their code blocks. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-27 01:20:14 +08:00
xcall_flush_tlb_page: /* 20 insns */
sparc64: Fix race in TLB batch processing. As reported by Dave Kleikamp, when we emit cross calls to do batched TLB flush processing we have a race because we do not synchronize on the sibling cpus completing the cross call. So meanwhile the TLB batch can be reset (tb->tlb_nr set to zero, etc.) and either flushes are missed or flushes will flush the wrong addresses. Fix this by using generic infrastructure to synchonize on the completion of the cross call. This first required getting the flush_tlb_pending() call out from switch_to() which operates with locks held and interrupts disabled. The problem is that smp_call_function_many() cannot be invoked with IRQs disabled and this is explicitly checked for with WARN_ON_ONCE(). We get the batch processing outside of locked IRQ disabled sections by using some ideas from the powerpc port. Namely, we only batch inside of arch_{enter,leave}_lazy_mmu_mode() calls. If we're not in such a region, we flush TLBs synchronously. 1) Get rid of xcall_flush_tlb_pending and per-cpu type implementations. 2) Do TLB batch cross calls instead via: smp_call_function_many() tlb_pending_func() __flush_tlb_pending() 3) Batch only in lazy mmu sequences: a) Add 'active' member to struct tlb_batch b) Define __HAVE_ARCH_ENTER_LAZY_MMU_MODE c) Set 'active' in arch_enter_lazy_mmu_mode() d) Run batch and clear 'active' in arch_leave_lazy_mmu_mode() e) Check 'active' in tlb_batch_add_one() and do a synchronous flush if it's clear. 4) Add infrastructure for synchronous TLB page flushes. a) Implement __flush_tlb_page and per-cpu variants, patch as needed. b) Likewise for xcall_flush_tlb_page. c) Implement smp_flush_tlb_page() to invoke the cross-call. d) Wire up global_flush_tlb_page() to the right routine based upon CONFIG_SMP 5) It turns out that singleton batches are very common, 2 out of every 3 batch flushes have only a single entry in them. The batch flush waiting is very expensive, both because of the poll on sibling cpu completeion, as well as because passing the tlb batch pointer to the sibling cpus invokes a shared memory dereference. Therefore, in flush_tlb_pending(), if there is only one entry in the batch perform a completely asynchronous global_flush_tlb_page() instead. Reported-by: Dave Kleikamp <dave.kleikamp@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Dave Kleikamp <dave.kleikamp@oracle.com>
2013-04-20 05:26:26 +08:00
/* %g5=context, %g1=vaddr */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
mov PRIMARY_CONTEXT, %g4
ldxa [%g4] ASI_DMMU, %g2
[SPARC64]: Preserve nucleus ctx page size during TLB flushes. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-08-31 11:21:34 +08:00
srlx %g2, CTX_PGSZ1_NUC_SHIFT, %g4
sllx %g4, CTX_PGSZ1_NUC_SHIFT, %g4
or %g5, %g4, %g5
mov PRIMARY_CONTEXT, %g4
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
stxa %g5, [%g4] ASI_DMMU
sparc64: Fix race in TLB batch processing. As reported by Dave Kleikamp, when we emit cross calls to do batched TLB flush processing we have a race because we do not synchronize on the sibling cpus completing the cross call. So meanwhile the TLB batch can be reset (tb->tlb_nr set to zero, etc.) and either flushes are missed or flushes will flush the wrong addresses. Fix this by using generic infrastructure to synchonize on the completion of the cross call. This first required getting the flush_tlb_pending() call out from switch_to() which operates with locks held and interrupts disabled. The problem is that smp_call_function_many() cannot be invoked with IRQs disabled and this is explicitly checked for with WARN_ON_ONCE(). We get the batch processing outside of locked IRQ disabled sections by using some ideas from the powerpc port. Namely, we only batch inside of arch_{enter,leave}_lazy_mmu_mode() calls. If we're not in such a region, we flush TLBs synchronously. 1) Get rid of xcall_flush_tlb_pending and per-cpu type implementations. 2) Do TLB batch cross calls instead via: smp_call_function_many() tlb_pending_func() __flush_tlb_pending() 3) Batch only in lazy mmu sequences: a) Add 'active' member to struct tlb_batch b) Define __HAVE_ARCH_ENTER_LAZY_MMU_MODE c) Set 'active' in arch_enter_lazy_mmu_mode() d) Run batch and clear 'active' in arch_leave_lazy_mmu_mode() e) Check 'active' in tlb_batch_add_one() and do a synchronous flush if it's clear. 4) Add infrastructure for synchronous TLB page flushes. a) Implement __flush_tlb_page and per-cpu variants, patch as needed. b) Likewise for xcall_flush_tlb_page. c) Implement smp_flush_tlb_page() to invoke the cross-call. d) Wire up global_flush_tlb_page() to the right routine based upon CONFIG_SMP 5) It turns out that singleton batches are very common, 2 out of every 3 batch flushes have only a single entry in them. The batch flush waiting is very expensive, both because of the poll on sibling cpu completeion, as well as because passing the tlb batch pointer to the sibling cpus invokes a shared memory dereference. Therefore, in flush_tlb_pending(), if there is only one entry in the batch perform a completely asynchronous global_flush_tlb_page() instead. Reported-by: Dave Kleikamp <dave.kleikamp@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Dave Kleikamp <dave.kleikamp@oracle.com>
2013-04-20 05:26:26 +08:00
andcc %g1, 0x1, %g0
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
be,pn %icc, 2f
sparc64: Fix race in TLB batch processing. As reported by Dave Kleikamp, when we emit cross calls to do batched TLB flush processing we have a race because we do not synchronize on the sibling cpus completing the cross call. So meanwhile the TLB batch can be reset (tb->tlb_nr set to zero, etc.) and either flushes are missed or flushes will flush the wrong addresses. Fix this by using generic infrastructure to synchonize on the completion of the cross call. This first required getting the flush_tlb_pending() call out from switch_to() which operates with locks held and interrupts disabled. The problem is that smp_call_function_many() cannot be invoked with IRQs disabled and this is explicitly checked for with WARN_ON_ONCE(). We get the batch processing outside of locked IRQ disabled sections by using some ideas from the powerpc port. Namely, we only batch inside of arch_{enter,leave}_lazy_mmu_mode() calls. If we're not in such a region, we flush TLBs synchronously. 1) Get rid of xcall_flush_tlb_pending and per-cpu type implementations. 2) Do TLB batch cross calls instead via: smp_call_function_many() tlb_pending_func() __flush_tlb_pending() 3) Batch only in lazy mmu sequences: a) Add 'active' member to struct tlb_batch b) Define __HAVE_ARCH_ENTER_LAZY_MMU_MODE c) Set 'active' in arch_enter_lazy_mmu_mode() d) Run batch and clear 'active' in arch_leave_lazy_mmu_mode() e) Check 'active' in tlb_batch_add_one() and do a synchronous flush if it's clear. 4) Add infrastructure for synchronous TLB page flushes. a) Implement __flush_tlb_page and per-cpu variants, patch as needed. b) Likewise for xcall_flush_tlb_page. c) Implement smp_flush_tlb_page() to invoke the cross-call. d) Wire up global_flush_tlb_page() to the right routine based upon CONFIG_SMP 5) It turns out that singleton batches are very common, 2 out of every 3 batch flushes have only a single entry in them. The batch flush waiting is very expensive, both because of the poll on sibling cpu completeion, as well as because passing the tlb batch pointer to the sibling cpus invokes a shared memory dereference. Therefore, in flush_tlb_pending(), if there is only one entry in the batch perform a completely asynchronous global_flush_tlb_page() instead. Reported-by: Dave Kleikamp <dave.kleikamp@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Dave Kleikamp <dave.kleikamp@oracle.com>
2013-04-20 05:26:26 +08:00
andn %g1, 0x1, %g5
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
stxa %g0, [%g5] ASI_IMMU_DEMAP
2: stxa %g0, [%g5] ASI_DMMU_DEMAP
membar #Sync
stxa %g2, [%g4] ASI_DMMU
retry
[SPARC64]: Bulletproof hypervisor TLB flushing. Check TLB flush hypervisor calls for errors and report them. Pass HV_MMU_ALL always for now, we can add back the optimization to avoid the I-TLB flush later. Always explicitly page align the virtual address arguments. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-27 11:31:49 +08:00
nop
sparc64: Fix race in TLB batch processing. As reported by Dave Kleikamp, when we emit cross calls to do batched TLB flush processing we have a race because we do not synchronize on the sibling cpus completing the cross call. So meanwhile the TLB batch can be reset (tb->tlb_nr set to zero, etc.) and either flushes are missed or flushes will flush the wrong addresses. Fix this by using generic infrastructure to synchonize on the completion of the cross call. This first required getting the flush_tlb_pending() call out from switch_to() which operates with locks held and interrupts disabled. The problem is that smp_call_function_many() cannot be invoked with IRQs disabled and this is explicitly checked for with WARN_ON_ONCE(). We get the batch processing outside of locked IRQ disabled sections by using some ideas from the powerpc port. Namely, we only batch inside of arch_{enter,leave}_lazy_mmu_mode() calls. If we're not in such a region, we flush TLBs synchronously. 1) Get rid of xcall_flush_tlb_pending and per-cpu type implementations. 2) Do TLB batch cross calls instead via: smp_call_function_many() tlb_pending_func() __flush_tlb_pending() 3) Batch only in lazy mmu sequences: a) Add 'active' member to struct tlb_batch b) Define __HAVE_ARCH_ENTER_LAZY_MMU_MODE c) Set 'active' in arch_enter_lazy_mmu_mode() d) Run batch and clear 'active' in arch_leave_lazy_mmu_mode() e) Check 'active' in tlb_batch_add_one() and do a synchronous flush if it's clear. 4) Add infrastructure for synchronous TLB page flushes. a) Implement __flush_tlb_page and per-cpu variants, patch as needed. b) Likewise for xcall_flush_tlb_page. c) Implement smp_flush_tlb_page() to invoke the cross-call. d) Wire up global_flush_tlb_page() to the right routine based upon CONFIG_SMP 5) It turns out that singleton batches are very common, 2 out of every 3 batch flushes have only a single entry in them. The batch flush waiting is very expensive, both because of the poll on sibling cpu completeion, as well as because passing the tlb batch pointer to the sibling cpus invokes a shared memory dereference. Therefore, in flush_tlb_pending(), if there is only one entry in the batch perform a completely asynchronous global_flush_tlb_page() instead. Reported-by: Dave Kleikamp <dave.kleikamp@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Dave Kleikamp <dave.kleikamp@oracle.com>
2013-04-20 05:26:26 +08:00
nop
sparc64: Fix illegal relative branches in hypervisor patched TLB cross-call code. Just like the non-cross-call TLB flush handlers, the cross-call ones need to avoid doing PC-relative branches outside of their code blocks. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-27 01:20:14 +08:00
nop
nop
nop
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
.globl xcall_flush_tlb_kernel_range
sparc64: Handle extremely large kernel TLB range flushes more gracefully. When the vmalloc area gets fragmented, and because the firmware mapping area sits between where modules live and the vmalloc area, we can sometimes receive requests for enormous kernel TLB range flushes. When this happens the cpu just spins flushing billions of pages and this triggers the NMI watchdog and other problems. We took care of this on the TSB side by doing a linear scan of the table once we pass a certain threshold. Do something similar for the TLB flush, however we are limited by the TLB flush facilities provided by the different chip variants. First of all we use an (mostly arbitrary) cut-off of 256K which is about 32 pages. This can be tuned in the future. The huge range code path for each chip works as follows: 1) On spitfire we flush all non-locked TLB entries using diagnostic acceses. 2) On cheetah we use the "flush all" TLB flush. 3) On sun4v/hypervisor we do a TLB context flush on context 0, which unlike previous chips does not remove "permanent" or locked entries. We could probably do something better on spitfire, such as limiting the flush to kernel TLB entries or even doing range comparisons. However that probably isn't worth it since those chips are old and the TLB only had 64 entries. Reported-by: James Clarke <jrtc27@jrtc27.com> Tested-by: James Clarke <jrtc27@jrtc27.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-28 00:04:54 +08:00
xcall_flush_tlb_kernel_range: /* 44 insns */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
sethi %hi(PAGE_SIZE - 1), %g2
or %g2, %lo(PAGE_SIZE - 1), %g2
andn %g1, %g2, %g1
andn %g7, %g2, %g7
sub %g7, %g1, %g3
sparc64: Handle extremely large kernel TLB range flushes more gracefully. When the vmalloc area gets fragmented, and because the firmware mapping area sits between where modules live and the vmalloc area, we can sometimes receive requests for enormous kernel TLB range flushes. When this happens the cpu just spins flushing billions of pages and this triggers the NMI watchdog and other problems. We took care of this on the TSB side by doing a linear scan of the table once we pass a certain threshold. Do something similar for the TLB flush, however we are limited by the TLB flush facilities provided by the different chip variants. First of all we use an (mostly arbitrary) cut-off of 256K which is about 32 pages. This can be tuned in the future. The huge range code path for each chip works as follows: 1) On spitfire we flush all non-locked TLB entries using diagnostic acceses. 2) On cheetah we use the "flush all" TLB flush. 3) On sun4v/hypervisor we do a TLB context flush on context 0, which unlike previous chips does not remove "permanent" or locked entries. We could probably do something better on spitfire, such as limiting the flush to kernel TLB entries or even doing range comparisons. However that probably isn't worth it since those chips are old and the TLB only had 64 entries. Reported-by: James Clarke <jrtc27@jrtc27.com> Tested-by: James Clarke <jrtc27@jrtc27.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-28 00:04:54 +08:00
srlx %g3, 18, %g2
brnz,pn %g2, 2f
add %g2, 1, %g2
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
sub %g3, %g2, %g3
or %g1, 0x20, %g1 ! Nucleus
1: stxa %g0, [%g1 + %g3] ASI_DMMU_DEMAP
stxa %g0, [%g1 + %g3] ASI_IMMU_DEMAP
membar #Sync
brnz,pt %g3, 1b
sub %g3, %g2, %g3
retry
sparc64: Handle extremely large kernel TLB range flushes more gracefully. When the vmalloc area gets fragmented, and because the firmware mapping area sits between where modules live and the vmalloc area, we can sometimes receive requests for enormous kernel TLB range flushes. When this happens the cpu just spins flushing billions of pages and this triggers the NMI watchdog and other problems. We took care of this on the TSB side by doing a linear scan of the table once we pass a certain threshold. Do something similar for the TLB flush, however we are limited by the TLB flush facilities provided by the different chip variants. First of all we use an (mostly arbitrary) cut-off of 256K which is about 32 pages. This can be tuned in the future. The huge range code path for each chip works as follows: 1) On spitfire we flush all non-locked TLB entries using diagnostic acceses. 2) On cheetah we use the "flush all" TLB flush. 3) On sun4v/hypervisor we do a TLB context flush on context 0, which unlike previous chips does not remove "permanent" or locked entries. We could probably do something better on spitfire, such as limiting the flush to kernel TLB entries or even doing range comparisons. However that probably isn't worth it since those chips are old and the TLB only had 64 entries. Reported-by: James Clarke <jrtc27@jrtc27.com> Tested-by: James Clarke <jrtc27@jrtc27.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-28 00:04:54 +08:00
2: mov 63 * 8, %g1
1: ldxa [%g1] ASI_ITLB_DATA_ACCESS, %g2
andcc %g2, 0x40, %g0 /* _PAGE_L_4U */
bne,pn %xcc, 2f
mov TLB_TAG_ACCESS, %g2
stxa %g0, [%g2] ASI_IMMU
stxa %g0, [%g1] ASI_ITLB_DATA_ACCESS
membar #Sync
2: ldxa [%g1] ASI_DTLB_DATA_ACCESS, %g2
andcc %g2, 0x40, %g0
bne,pn %xcc, 2f
mov TLB_TAG_ACCESS, %g2
stxa %g0, [%g2] ASI_DMMU
stxa %g0, [%g1] ASI_DTLB_DATA_ACCESS
membar #Sync
2: sub %g1, 8, %g1
brgez,pt %g1, 1b
nop
retry
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
nop
nop
nop
[SPARC64]: Bulletproof hypervisor TLB flushing. Check TLB flush hypervisor calls for errors and report them. Pass HV_MMU_ALL always for now, we can add back the optimization to avoid the I-TLB flush later. Always explicitly page align the virtual address arguments. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-27 11:31:49 +08:00
nop
nop
nop
sparc64: Fix illegal relative branches in hypervisor patched TLB cross-call code. Just like the non-cross-call TLB flush handlers, the cross-call ones need to avoid doing PC-relative branches outside of their code blocks. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-27 01:20:14 +08:00
nop
nop
nop
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* This runs in a very controlled environment, so we do
* not need to worry about BH races etc.
*/
.globl xcall_sync_tick
xcall_sync_tick:
[SPARC64]: Sanitize %pstate writes for sun4v. If we're just switching between different alternate global sets, nop it out on sun4v. Also, get rid of all of the alternate global save/restore in the OBP CIF trampoline code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-06 14:27:28 +08:00
661: rdpr %pstate, %g2
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
wrpr %g2, PSTATE_IG | PSTATE_AG, %pstate
[SPARC64]: Rename gl_{1,2}insn_patch --> sun4v_{1,2}insn_patch Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-07 16:00:16 +08:00
.section .sun4v_2insn_patch, "ax"
[SPARC64]: Sanitize %pstate writes for sun4v. If we're just switching between different alternate global sets, nop it out on sun4v. Also, get rid of all of the alternate global save/restore in the OBP CIF trampoline code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-06 14:27:28 +08:00
.word 661b
nop
nop
.previous
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
rdpr %pil, %g2
sparc64: Make %pil level 15 a pseudo-NMI. So that we can profile code even in a local_irq_disable() section, only write 14 (instead of 15) into the %pil register to disable IRQs. This allows PIL level 15 to serve as a pseudo NMI. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-24 13:55:29 +08:00
wrpr %g0, PIL_NORMAL_MAX, %pil
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
sethi %hi(109f), %g7
b,pt %xcc, etrap_irq
109: or %g7, %lo(109b), %g7
[SPARC64]: Add irqtrace/stacktrace/lockdep support. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-11-17 05:38:57 +08:00
#ifdef CONFIG_TRACE_IRQFLAGS
call trace_hardirqs_off
nop
#endif
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
call smp_synchronize_tick_client
nop
b rtrap_xcall
ldx [%sp + PTREGS_OFF + PT_V9_TSTATE], %l1
sparc64: Add global register dumping facility. When a cpu really is stuck in the kernel, it can be often impossible to figure out which cpu is stuck where. The worst case is when the stuck cpu has interrupts disabled. Therefore, implement a global cpu state capture that uses SMP message interrupts which are not disabled by the normal IRQ enable/disable APIs of the kernel. As long as we can get a sysrq 'y' to the kernel, we can get a dump. Even if the console interrupt cpu is wedged, we can trigger it from userspace using /proc/sysrq-trigger The output is made compact so that this facility is more useful on high cpu count systems, which is where this facility will likely find itself the most useful :) Signed-off-by: David S. Miller <davem@davemloft.net>
2008-05-20 14:46:00 +08:00
.globl xcall_fetch_glob_regs
xcall_fetch_glob_regs:
sparc64: Add global PMU register dumping via sysrq. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-10-17 00:34:01 +08:00
sethi %hi(global_cpu_snapshot), %g1
or %g1, %lo(global_cpu_snapshot), %g1
sparc64: Add global register dumping facility. When a cpu really is stuck in the kernel, it can be often impossible to figure out which cpu is stuck where. The worst case is when the stuck cpu has interrupts disabled. Therefore, implement a global cpu state capture that uses SMP message interrupts which are not disabled by the normal IRQ enable/disable APIs of the kernel. As long as we can get a sysrq 'y' to the kernel, we can get a dump. Even if the console interrupt cpu is wedged, we can trigger it from userspace using /proc/sysrq-trigger The output is made compact so that this facility is more useful on high cpu count systems, which is where this facility will likely find itself the most useful :) Signed-off-by: David S. Miller <davem@davemloft.net>
2008-05-20 14:46:00 +08:00
__GET_CPUID(%g2)
sllx %g2, 6, %g3
add %g1, %g3, %g1
rdpr %tstate, %g7
stx %g7, [%g1 + GR_SNAP_TSTATE]
rdpr %tpc, %g7
stx %g7, [%g1 + GR_SNAP_TPC]
rdpr %tnpc, %g7
stx %g7, [%g1 + GR_SNAP_TNPC]
stx %o7, [%g1 + GR_SNAP_O7]
stx %i7, [%g1 + GR_SNAP_I7]
sparc64: Make global reg dumping even more useful. Record one more level of stack frame program counter. Particularly when lockdep and all sorts of spinlock debugging is enabled, figuring out the caller of spin_lock() is difficult when the cpu is stuck on the lock. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-31 12:57:59 +08:00
/* Don't try this at home kids... */
sparc64: Do not clobber %g2 in xcall_fetch_glob_regs(). %g2 is meant to hold the CPUID number throughout this routine, since at the very beginning, and at the very end, we use %g2 to calculate indexes into per-cpu arrays. However we erroneously clobber it in order to hold the %cwp register value mid-stream. Fix this code to use %g3 for the %cwp read and related calulcations instead. Reported-by: Meelis Roos <mroos@linux.ee> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-05-11 02:00:46 +08:00
rdpr %cwp, %g3
sub %g3, 1, %g7
sparc64: Make global reg dumping even more useful. Record one more level of stack frame program counter. Particularly when lockdep and all sorts of spinlock debugging is enabled, figuring out the caller of spin_lock() is difficult when the cpu is stuck on the lock. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-31 12:57:59 +08:00
wrpr %g7, %cwp
mov %i7, %g7
sparc64: Do not clobber %g2 in xcall_fetch_glob_regs(). %g2 is meant to hold the CPUID number throughout this routine, since at the very beginning, and at the very end, we use %g2 to calculate indexes into per-cpu arrays. However we erroneously clobber it in order to hold the %cwp register value mid-stream. Fix this code to use %g3 for the %cwp read and related calulcations instead. Reported-by: Meelis Roos <mroos@linux.ee> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-05-11 02:00:46 +08:00
wrpr %g3, %cwp
sparc64: Make global reg dumping even more useful. Record one more level of stack frame program counter. Particularly when lockdep and all sorts of spinlock debugging is enabled, figuring out the caller of spin_lock() is difficult when the cpu is stuck on the lock. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-31 12:57:59 +08:00
stx %g7, [%g1 + GR_SNAP_RPC]
sparc64: Add global register dumping facility. When a cpu really is stuck in the kernel, it can be often impossible to figure out which cpu is stuck where. The worst case is when the stuck cpu has interrupts disabled. Therefore, implement a global cpu state capture that uses SMP message interrupts which are not disabled by the normal IRQ enable/disable APIs of the kernel. As long as we can get a sysrq 'y' to the kernel, we can get a dump. Even if the console interrupt cpu is wedged, we can trigger it from userspace using /proc/sysrq-trigger The output is made compact so that this facility is more useful on high cpu count systems, which is where this facility will likely find itself the most useful :) Signed-off-by: David S. Miller <davem@davemloft.net>
2008-05-20 14:46:00 +08:00
sethi %hi(trap_block), %g7
or %g7, %lo(trap_block), %g7
sllx %g2, TRAP_BLOCK_SZ_SHIFT, %g2
add %g7, %g2, %g7
ldx [%g7 + TRAP_PER_CPU_THREAD], %g3
stx %g3, [%g1 + GR_SNAP_THREAD]
retry
sparc64: Add global PMU register dumping via sysrq. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-10-17 00:34:01 +08:00
.globl xcall_fetch_glob_pmu
xcall_fetch_glob_pmu:
sethi %hi(global_cpu_snapshot), %g1
or %g1, %lo(global_cpu_snapshot), %g1
__GET_CPUID(%g2)
sllx %g2, 6, %g3
add %g1, %g3, %g1
rd %pic, %g7
stx %g7, [%g1 + (4 * 8)]
rd %pcr, %g7
stx %g7, [%g1 + (0 * 8)]
retry
.globl xcall_fetch_glob_pmu_n4
xcall_fetch_glob_pmu_n4:
sethi %hi(global_cpu_snapshot), %g1
or %g1, %lo(global_cpu_snapshot), %g1
__GET_CPUID(%g2)
sllx %g2, 6, %g3
add %g1, %g3, %g1
ldxa [%g0] ASI_PIC, %g7
stx %g7, [%g1 + (4 * 8)]
mov 0x08, %g3
ldxa [%g3] ASI_PIC, %g7
stx %g7, [%g1 + (5 * 8)]
mov 0x10, %g3
ldxa [%g3] ASI_PIC, %g7
stx %g7, [%g1 + (6 * 8)]
mov 0x18, %g3
ldxa [%g3] ASI_PIC, %g7
stx %g7, [%g1 + (7 * 8)]
mov %o0, %g2
mov %o1, %g3
mov %o5, %g7
mov HV_FAST_VT_GET_PERFREG, %o5
mov 3, %o0
ta HV_FAST_TRAP
stx %o1, [%g1 + (3 * 8)]
mov HV_FAST_VT_GET_PERFREG, %o5
mov 2, %o0
ta HV_FAST_TRAP
stx %o1, [%g1 + (2 * 8)]
mov HV_FAST_VT_GET_PERFREG, %o5
mov 1, %o0
ta HV_FAST_TRAP
stx %o1, [%g1 + (1 * 8)]
mov HV_FAST_VT_GET_PERFREG, %o5
mov 0, %o0
ta HV_FAST_TRAP
stx %o1, [%g1 + (0 * 8)]
mov %g2, %o0
mov %g3, %o1
mov %g7, %o5
retry
sparc64: Handle extremely large kernel TLB range flushes more gracefully. When the vmalloc area gets fragmented, and because the firmware mapping area sits between where modules live and the vmalloc area, we can sometimes receive requests for enormous kernel TLB range flushes. When this happens the cpu just spins flushing billions of pages and this triggers the NMI watchdog and other problems. We took care of this on the TSB side by doing a linear scan of the table once we pass a certain threshold. Do something similar for the TLB flush, however we are limited by the TLB flush facilities provided by the different chip variants. First of all we use an (mostly arbitrary) cut-off of 256K which is about 32 pages. This can be tuned in the future. The huge range code path for each chip works as follows: 1) On spitfire we flush all non-locked TLB entries using diagnostic acceses. 2) On cheetah we use the "flush all" TLB flush. 3) On sun4v/hypervisor we do a TLB context flush on context 0, which unlike previous chips does not remove "permanent" or locked entries. We could probably do something better on spitfire, such as limiting the flush to kernel TLB entries or even doing range comparisons. However that probably isn't worth it since those chips are old and the TLB only had 64 entries. Reported-by: James Clarke <jrtc27@jrtc27.com> Tested-by: James Clarke <jrtc27@jrtc27.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-28 00:04:54 +08:00
__cheetah_xcall_flush_tlb_kernel_range: /* 44 insns */
sethi %hi(PAGE_SIZE - 1), %g2
or %g2, %lo(PAGE_SIZE - 1), %g2
andn %g1, %g2, %g1
andn %g7, %g2, %g7
sub %g7, %g1, %g3
srlx %g3, 18, %g2
brnz,pn %g2, 2f
add %g2, 1, %g2
sub %g3, %g2, %g3
or %g1, 0x20, %g1 ! Nucleus
1: stxa %g0, [%g1 + %g3] ASI_DMMU_DEMAP
stxa %g0, [%g1 + %g3] ASI_IMMU_DEMAP
membar #Sync
brnz,pt %g3, 1b
sub %g3, %g2, %g3
retry
2: mov 0x80, %g2
stxa %g0, [%g2] ASI_DMMU_DEMAP
membar #Sync
stxa %g0, [%g2] ASI_IMMU_DEMAP
membar #Sync
retry
nop
nop
nop
nop
nop
nop
nop
nop
nop
nop
nop
nop
nop
nop
nop
nop
nop
nop
nop
nop
nop
nop
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
#ifdef DCACHE_ALIASING_POSSIBLE
.align 32
.globl xcall_flush_dcache_page_cheetah
xcall_flush_dcache_page_cheetah: /* %g1 == physical page address */
sethi %hi(PAGE_SIZE), %g3
1: subcc %g3, (1 << 5), %g3
stxa %g0, [%g1 + %g3] ASI_DCACHE_INVALIDATE
membar #Sync
bne,pt %icc, 1b
nop
retry
nop
#endif /* DCACHE_ALIASING_POSSIBLE */
.globl xcall_flush_dcache_page_spitfire
xcall_flush_dcache_page_spitfire: /* %g1 == physical page address
%g7 == kernel page virtual address
%g5 == (page->mapping != NULL) */
#ifdef DCACHE_ALIASING_POSSIBLE
srlx %g1, (13 - 2), %g1 ! Form tag comparitor
sethi %hi(L1DCACHE_SIZE), %g3 ! D$ size == 16K
sub %g3, (1 << 5), %g3 ! D$ linesize == 32
1: ldxa [%g3] ASI_DCACHE_TAG, %g2
andcc %g2, 0x3, %g0
be,pn %xcc, 2f
andn %g2, 0x3, %g2
cmp %g2, %g1
bne,pt %xcc, 2f
nop
stxa %g0, [%g3] ASI_DCACHE_TAG
membar #Sync
2: cmp %g3, 0
bne,pt %xcc, 1b
sub %g3, (1 << 5), %g3
brz,pn %g5, 2f
#endif /* DCACHE_ALIASING_POSSIBLE */
sethi %hi(PAGE_SIZE), %g3
1: flush %g7
subcc %g3, (1 << 5), %g3
bne,pt %icc, 1b
add %g7, (1 << 5), %g7
2: retry
nop
nop
[SPARC64]: Bulletproof hypervisor TLB flushing. Check TLB flush hypervisor calls for errors and report them. Pass HV_MMU_ALL always for now, we can add back the optimization to avoid the I-TLB flush later. Always explicitly page align the virtual address arguments. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-27 11:31:49 +08:00
/* %g5: error
* %g6: tlb op
*/
__hypervisor_tlb_xcall_error:
mov %g5, %g4
mov %g6, %g5
ba,pt %xcc, etrap
rd %pc, %g7
mov %l4, %o0
call hypervisor_tlbop_error_xcall
mov %l5, %o1
[SPARC64]: %l6 trap return handling no longer necessary. Now that we indicate the "restart system call" in the trap type field of pt_regs->magic, we don't need to set the %l6 boolean in all of the trap return paths. And we therefore don't need to pass it to do_notify_resume(). Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-24 18:15:22 +08:00
ba,a,pt %xcc, rtrap
[SPARC64]: Bulletproof hypervisor TLB flushing. Check TLB flush hypervisor calls for errors and report them. Pass HV_MMU_ALL always for now, we can add back the optimization to avoid the I-TLB flush later. Always explicitly page align the virtual address arguments. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-27 11:31:49 +08:00
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
.globl __hypervisor_xcall_flush_tlb_mm
sparc64: Fix illegal relative branches in hypervisor patched TLB cross-call code. Just like the non-cross-call TLB flush handlers, the cross-call ones need to avoid doing PC-relative branches outside of their code blocks. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-27 01:20:14 +08:00
__hypervisor_xcall_flush_tlb_mm: /* 24 insns */
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
/* %g5=ctx, g1,g2,g3,g4,g7=scratch, %g6=unusable */
mov %o0, %g2
mov %o1, %g3
mov %o2, %g4
mov %o3, %g1
mov %o5, %g7
clr %o0 /* ARG0: CPU lists unimplemented */
clr %o1 /* ARG1: CPU lists unimplemented */
mov %g5, %o2 /* ARG2: mmu context */
mov HV_MMU_ALL, %o3 /* ARG3: flags */
mov HV_FAST_MMU_DEMAP_CTX, %o5
ta HV_FAST_TRAP
[SPARC64]: Bulletproof hypervisor TLB flushing. Check TLB flush hypervisor calls for errors and report them. Pass HV_MMU_ALL always for now, we can add back the optimization to avoid the I-TLB flush later. Always explicitly page align the virtual address arguments. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-27 11:31:49 +08:00
mov HV_FAST_MMU_DEMAP_CTX, %g6
sparc64: Fix illegal relative branches in hypervisor patched TLB cross-call code. Just like the non-cross-call TLB flush handlers, the cross-call ones need to avoid doing PC-relative branches outside of their code blocks. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-27 01:20:14 +08:00
brnz,pn %o0, 1f
[SPARC64]: Bulletproof hypervisor TLB flushing. Check TLB flush hypervisor calls for errors and report them. Pass HV_MMU_ALL always for now, we can add back the optimization to avoid the I-TLB flush later. Always explicitly page align the virtual address arguments. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-27 11:31:49 +08:00
mov %o0, %g5
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
mov %g2, %o0
mov %g3, %o1
mov %g4, %o2
mov %g1, %o3
mov %g7, %o5
membar #Sync
retry
sparc64: Fix illegal relative branches in hypervisor patched TLB cross-call code. Just like the non-cross-call TLB flush handlers, the cross-call ones need to avoid doing PC-relative branches outside of their code blocks. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-27 01:20:14 +08:00
1: sethi %hi(__hypervisor_tlb_xcall_error), %g4
jmpl %g4 + %lo(__hypervisor_tlb_xcall_error), %g0
nop
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
sparc64: Fix race in TLB batch processing. As reported by Dave Kleikamp, when we emit cross calls to do batched TLB flush processing we have a race because we do not synchronize on the sibling cpus completing the cross call. So meanwhile the TLB batch can be reset (tb->tlb_nr set to zero, etc.) and either flushes are missed or flushes will flush the wrong addresses. Fix this by using generic infrastructure to synchonize on the completion of the cross call. This first required getting the flush_tlb_pending() call out from switch_to() which operates with locks held and interrupts disabled. The problem is that smp_call_function_many() cannot be invoked with IRQs disabled and this is explicitly checked for with WARN_ON_ONCE(). We get the batch processing outside of locked IRQ disabled sections by using some ideas from the powerpc port. Namely, we only batch inside of arch_{enter,leave}_lazy_mmu_mode() calls. If we're not in such a region, we flush TLBs synchronously. 1) Get rid of xcall_flush_tlb_pending and per-cpu type implementations. 2) Do TLB batch cross calls instead via: smp_call_function_many() tlb_pending_func() __flush_tlb_pending() 3) Batch only in lazy mmu sequences: a) Add 'active' member to struct tlb_batch b) Define __HAVE_ARCH_ENTER_LAZY_MMU_MODE c) Set 'active' in arch_enter_lazy_mmu_mode() d) Run batch and clear 'active' in arch_leave_lazy_mmu_mode() e) Check 'active' in tlb_batch_add_one() and do a synchronous flush if it's clear. 4) Add infrastructure for synchronous TLB page flushes. a) Implement __flush_tlb_page and per-cpu variants, patch as needed. b) Likewise for xcall_flush_tlb_page. c) Implement smp_flush_tlb_page() to invoke the cross-call. d) Wire up global_flush_tlb_page() to the right routine based upon CONFIG_SMP 5) It turns out that singleton batches are very common, 2 out of every 3 batch flushes have only a single entry in them. The batch flush waiting is very expensive, both because of the poll on sibling cpu completeion, as well as because passing the tlb batch pointer to the sibling cpus invokes a shared memory dereference. Therefore, in flush_tlb_pending(), if there is only one entry in the batch perform a completely asynchronous global_flush_tlb_page() instead. Reported-by: Dave Kleikamp <dave.kleikamp@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Dave Kleikamp <dave.kleikamp@oracle.com>
2013-04-20 05:26:26 +08:00
.globl __hypervisor_xcall_flush_tlb_page
sparc64: Fix illegal relative branches in hypervisor patched TLB cross-call code. Just like the non-cross-call TLB flush handlers, the cross-call ones need to avoid doing PC-relative branches outside of their code blocks. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-27 01:20:14 +08:00
__hypervisor_xcall_flush_tlb_page: /* 20 insns */
sparc64: Fix race in TLB batch processing. As reported by Dave Kleikamp, when we emit cross calls to do batched TLB flush processing we have a race because we do not synchronize on the sibling cpus completing the cross call. So meanwhile the TLB batch can be reset (tb->tlb_nr set to zero, etc.) and either flushes are missed or flushes will flush the wrong addresses. Fix this by using generic infrastructure to synchonize on the completion of the cross call. This first required getting the flush_tlb_pending() call out from switch_to() which operates with locks held and interrupts disabled. The problem is that smp_call_function_many() cannot be invoked with IRQs disabled and this is explicitly checked for with WARN_ON_ONCE(). We get the batch processing outside of locked IRQ disabled sections by using some ideas from the powerpc port. Namely, we only batch inside of arch_{enter,leave}_lazy_mmu_mode() calls. If we're not in such a region, we flush TLBs synchronously. 1) Get rid of xcall_flush_tlb_pending and per-cpu type implementations. 2) Do TLB batch cross calls instead via: smp_call_function_many() tlb_pending_func() __flush_tlb_pending() 3) Batch only in lazy mmu sequences: a) Add 'active' member to struct tlb_batch b) Define __HAVE_ARCH_ENTER_LAZY_MMU_MODE c) Set 'active' in arch_enter_lazy_mmu_mode() d) Run batch and clear 'active' in arch_leave_lazy_mmu_mode() e) Check 'active' in tlb_batch_add_one() and do a synchronous flush if it's clear. 4) Add infrastructure for synchronous TLB page flushes. a) Implement __flush_tlb_page and per-cpu variants, patch as needed. b) Likewise for xcall_flush_tlb_page. c) Implement smp_flush_tlb_page() to invoke the cross-call. d) Wire up global_flush_tlb_page() to the right routine based upon CONFIG_SMP 5) It turns out that singleton batches are very common, 2 out of every 3 batch flushes have only a single entry in them. The batch flush waiting is very expensive, both because of the poll on sibling cpu completeion, as well as because passing the tlb batch pointer to the sibling cpus invokes a shared memory dereference. Therefore, in flush_tlb_pending(), if there is only one entry in the batch perform a completely asynchronous global_flush_tlb_page() instead. Reported-by: Dave Kleikamp <dave.kleikamp@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Dave Kleikamp <dave.kleikamp@oracle.com>
2013-04-20 05:26:26 +08:00
/* %g5=ctx, %g1=vaddr */
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
mov %o0, %g2
mov %o1, %g3
mov %o2, %g4
sparc64: Fix race in TLB batch processing. As reported by Dave Kleikamp, when we emit cross calls to do batched TLB flush processing we have a race because we do not synchronize on the sibling cpus completing the cross call. So meanwhile the TLB batch can be reset (tb->tlb_nr set to zero, etc.) and either flushes are missed or flushes will flush the wrong addresses. Fix this by using generic infrastructure to synchonize on the completion of the cross call. This first required getting the flush_tlb_pending() call out from switch_to() which operates with locks held and interrupts disabled. The problem is that smp_call_function_many() cannot be invoked with IRQs disabled and this is explicitly checked for with WARN_ON_ONCE(). We get the batch processing outside of locked IRQ disabled sections by using some ideas from the powerpc port. Namely, we only batch inside of arch_{enter,leave}_lazy_mmu_mode() calls. If we're not in such a region, we flush TLBs synchronously. 1) Get rid of xcall_flush_tlb_pending and per-cpu type implementations. 2) Do TLB batch cross calls instead via: smp_call_function_many() tlb_pending_func() __flush_tlb_pending() 3) Batch only in lazy mmu sequences: a) Add 'active' member to struct tlb_batch b) Define __HAVE_ARCH_ENTER_LAZY_MMU_MODE c) Set 'active' in arch_enter_lazy_mmu_mode() d) Run batch and clear 'active' in arch_leave_lazy_mmu_mode() e) Check 'active' in tlb_batch_add_one() and do a synchronous flush if it's clear. 4) Add infrastructure for synchronous TLB page flushes. a) Implement __flush_tlb_page and per-cpu variants, patch as needed. b) Likewise for xcall_flush_tlb_page. c) Implement smp_flush_tlb_page() to invoke the cross-call. d) Wire up global_flush_tlb_page() to the right routine based upon CONFIG_SMP 5) It turns out that singleton batches are very common, 2 out of every 3 batch flushes have only a single entry in them. The batch flush waiting is very expensive, both because of the poll on sibling cpu completeion, as well as because passing the tlb batch pointer to the sibling cpus invokes a shared memory dereference. Therefore, in flush_tlb_pending(), if there is only one entry in the batch perform a completely asynchronous global_flush_tlb_page() instead. Reported-by: Dave Kleikamp <dave.kleikamp@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Dave Kleikamp <dave.kleikamp@oracle.com>
2013-04-20 05:26:26 +08:00
mov %g1, %o0 /* ARG0: virtual address */
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
mov %g5, %o1 /* ARG1: mmu context */
[SPARC64]: Bulletproof hypervisor TLB flushing. Check TLB flush hypervisor calls for errors and report them. Pass HV_MMU_ALL always for now, we can add back the optimization to avoid the I-TLB flush later. Always explicitly page align the virtual address arguments. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-27 11:31:49 +08:00
mov HV_MMU_ALL, %o2 /* ARG2: flags */
srlx %o0, PAGE_SHIFT, %o0
sllx %o0, PAGE_SHIFT, %o0
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
ta HV_MMU_UNMAP_ADDR_TRAP
[SPARC64]: Bulletproof hypervisor TLB flushing. Check TLB flush hypervisor calls for errors and report them. Pass HV_MMU_ALL always for now, we can add back the optimization to avoid the I-TLB flush later. Always explicitly page align the virtual address arguments. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-27 11:31:49 +08:00
mov HV_MMU_UNMAP_ADDR_TRAP, %g6
sparc64: Fix illegal relative branches in hypervisor patched TLB cross-call code. Just like the non-cross-call TLB flush handlers, the cross-call ones need to avoid doing PC-relative branches outside of their code blocks. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-27 01:20:14 +08:00
brnz,a,pn %o0, 1f
[SPARC64]: Bulletproof hypervisor TLB flushing. Check TLB flush hypervisor calls for errors and report them. Pass HV_MMU_ALL always for now, we can add back the optimization to avoid the I-TLB flush later. Always explicitly page align the virtual address arguments. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-27 11:31:49 +08:00
mov %o0, %g5
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
mov %g2, %o0
mov %g3, %o1
mov %g4, %o2
membar #Sync
retry
sparc64: Fix illegal relative branches in hypervisor patched TLB cross-call code. Just like the non-cross-call TLB flush handlers, the cross-call ones need to avoid doing PC-relative branches outside of their code blocks. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-27 01:20:14 +08:00
1: sethi %hi(__hypervisor_tlb_xcall_error), %g4
jmpl %g4 + %lo(__hypervisor_tlb_xcall_error), %g0
nop
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
.globl __hypervisor_xcall_flush_tlb_kernel_range
sparc64: Handle extremely large kernel TLB range flushes more gracefully. When the vmalloc area gets fragmented, and because the firmware mapping area sits between where modules live and the vmalloc area, we can sometimes receive requests for enormous kernel TLB range flushes. When this happens the cpu just spins flushing billions of pages and this triggers the NMI watchdog and other problems. We took care of this on the TSB side by doing a linear scan of the table once we pass a certain threshold. Do something similar for the TLB flush, however we are limited by the TLB flush facilities provided by the different chip variants. First of all we use an (mostly arbitrary) cut-off of 256K which is about 32 pages. This can be tuned in the future. The huge range code path for each chip works as follows: 1) On spitfire we flush all non-locked TLB entries using diagnostic acceses. 2) On cheetah we use the "flush all" TLB flush. 3) On sun4v/hypervisor we do a TLB context flush on context 0, which unlike previous chips does not remove "permanent" or locked entries. We could probably do something better on spitfire, such as limiting the flush to kernel TLB entries or even doing range comparisons. However that probably isn't worth it since those chips are old and the TLB only had 64 entries. Reported-by: James Clarke <jrtc27@jrtc27.com> Tested-by: James Clarke <jrtc27@jrtc27.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-28 00:04:54 +08:00
__hypervisor_xcall_flush_tlb_kernel_range: /* 44 insns */
[SPARC64]: Bulletproof hypervisor TLB flushing. Check TLB flush hypervisor calls for errors and report them. Pass HV_MMU_ALL always for now, we can add back the optimization to avoid the I-TLB flush later. Always explicitly page align the virtual address arguments. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-27 11:31:49 +08:00
/* %g1=start, %g7=end, g2,g3,g4,g5,g6=scratch */
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
sethi %hi(PAGE_SIZE - 1), %g2
or %g2, %lo(PAGE_SIZE - 1), %g2
andn %g1, %g2, %g1
andn %g7, %g2, %g7
sub %g7, %g1, %g3
sparc64: Handle extremely large kernel TLB range flushes more gracefully. When the vmalloc area gets fragmented, and because the firmware mapping area sits between where modules live and the vmalloc area, we can sometimes receive requests for enormous kernel TLB range flushes. When this happens the cpu just spins flushing billions of pages and this triggers the NMI watchdog and other problems. We took care of this on the TSB side by doing a linear scan of the table once we pass a certain threshold. Do something similar for the TLB flush, however we are limited by the TLB flush facilities provided by the different chip variants. First of all we use an (mostly arbitrary) cut-off of 256K which is about 32 pages. This can be tuned in the future. The huge range code path for each chip works as follows: 1) On spitfire we flush all non-locked TLB entries using diagnostic acceses. 2) On cheetah we use the "flush all" TLB flush. 3) On sun4v/hypervisor we do a TLB context flush on context 0, which unlike previous chips does not remove "permanent" or locked entries. We could probably do something better on spitfire, such as limiting the flush to kernel TLB entries or even doing range comparisons. However that probably isn't worth it since those chips are old and the TLB only had 64 entries. Reported-by: James Clarke <jrtc27@jrtc27.com> Tested-by: James Clarke <jrtc27@jrtc27.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-28 00:04:54 +08:00
srlx %g3, 18, %g7
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
add %g2, 1, %g2
sub %g3, %g2, %g3
mov %o0, %g2
mov %o1, %g4
sparc64: Handle extremely large kernel TLB range flushes more gracefully. When the vmalloc area gets fragmented, and because the firmware mapping area sits between where modules live and the vmalloc area, we can sometimes receive requests for enormous kernel TLB range flushes. When this happens the cpu just spins flushing billions of pages and this triggers the NMI watchdog and other problems. We took care of this on the TSB side by doing a linear scan of the table once we pass a certain threshold. Do something similar for the TLB flush, however we are limited by the TLB flush facilities provided by the different chip variants. First of all we use an (mostly arbitrary) cut-off of 256K which is about 32 pages. This can be tuned in the future. The huge range code path for each chip works as follows: 1) On spitfire we flush all non-locked TLB entries using diagnostic acceses. 2) On cheetah we use the "flush all" TLB flush. 3) On sun4v/hypervisor we do a TLB context flush on context 0, which unlike previous chips does not remove "permanent" or locked entries. We could probably do something better on spitfire, such as limiting the flush to kernel TLB entries or even doing range comparisons. However that probably isn't worth it since those chips are old and the TLB only had 64 entries. Reported-by: James Clarke <jrtc27@jrtc27.com> Tested-by: James Clarke <jrtc27@jrtc27.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-28 00:04:54 +08:00
brnz,pn %g7, 2f
mov %o2, %g7
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
1: add %g1, %g3, %o0 /* ARG0: virtual address */
mov 0, %o1 /* ARG1: mmu context */
mov HV_MMU_ALL, %o2 /* ARG2: flags */
ta HV_MMU_UNMAP_ADDR_TRAP
[SPARC64]: Bulletproof hypervisor TLB flushing. Check TLB flush hypervisor calls for errors and report them. Pass HV_MMU_ALL always for now, we can add back the optimization to avoid the I-TLB flush later. Always explicitly page align the virtual address arguments. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-27 11:31:49 +08:00
mov HV_MMU_UNMAP_ADDR_TRAP, %g6
sparc64: Fix illegal relative branches in hypervisor patched TLB cross-call code. Just like the non-cross-call TLB flush handlers, the cross-call ones need to avoid doing PC-relative branches outside of their code blocks. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-27 01:20:14 +08:00
brnz,pn %o0, 1f
[SPARC64]: Bulletproof hypervisor TLB flushing. Check TLB flush hypervisor calls for errors and report them. Pass HV_MMU_ALL always for now, we can add back the optimization to avoid the I-TLB flush later. Always explicitly page align the virtual address arguments. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-27 11:31:49 +08:00
mov %o0, %g5
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
sethi %hi(PAGE_SIZE), %o2
brnz,pt %g3, 1b
sub %g3, %o2, %g3
sparc64: Handle extremely large kernel TLB range flushes more gracefully. When the vmalloc area gets fragmented, and because the firmware mapping area sits between where modules live and the vmalloc area, we can sometimes receive requests for enormous kernel TLB range flushes. When this happens the cpu just spins flushing billions of pages and this triggers the NMI watchdog and other problems. We took care of this on the TSB side by doing a linear scan of the table once we pass a certain threshold. Do something similar for the TLB flush, however we are limited by the TLB flush facilities provided by the different chip variants. First of all we use an (mostly arbitrary) cut-off of 256K which is about 32 pages. This can be tuned in the future. The huge range code path for each chip works as follows: 1) On spitfire we flush all non-locked TLB entries using diagnostic acceses. 2) On cheetah we use the "flush all" TLB flush. 3) On sun4v/hypervisor we do a TLB context flush on context 0, which unlike previous chips does not remove "permanent" or locked entries. We could probably do something better on spitfire, such as limiting the flush to kernel TLB entries or even doing range comparisons. However that probably isn't worth it since those chips are old and the TLB only had 64 entries. Reported-by: James Clarke <jrtc27@jrtc27.com> Tested-by: James Clarke <jrtc27@jrtc27.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-28 00:04:54 +08:00
5: mov %g2, %o0
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
mov %g4, %o1
[SPARC64]: Bulletproof hypervisor TLB flushing. Check TLB flush hypervisor calls for errors and report them. Pass HV_MMU_ALL always for now, we can add back the optimization to avoid the I-TLB flush later. Always explicitly page align the virtual address arguments. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-27 11:31:49 +08:00
mov %g7, %o2
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
membar #Sync
retry
sparc64: Fix illegal relative branches in hypervisor patched TLB cross-call code. Just like the non-cross-call TLB flush handlers, the cross-call ones need to avoid doing PC-relative branches outside of their code blocks. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-27 01:20:14 +08:00
1: sethi %hi(__hypervisor_tlb_xcall_error), %g4
jmpl %g4 + %lo(__hypervisor_tlb_xcall_error), %g0
nop
sparc64: Handle extremely large kernel TLB range flushes more gracefully. When the vmalloc area gets fragmented, and because the firmware mapping area sits between where modules live and the vmalloc area, we can sometimes receive requests for enormous kernel TLB range flushes. When this happens the cpu just spins flushing billions of pages and this triggers the NMI watchdog and other problems. We took care of this on the TSB side by doing a linear scan of the table once we pass a certain threshold. Do something similar for the TLB flush, however we are limited by the TLB flush facilities provided by the different chip variants. First of all we use an (mostly arbitrary) cut-off of 256K which is about 32 pages. This can be tuned in the future. The huge range code path for each chip works as follows: 1) On spitfire we flush all non-locked TLB entries using diagnostic acceses. 2) On cheetah we use the "flush all" TLB flush. 3) On sun4v/hypervisor we do a TLB context flush on context 0, which unlike previous chips does not remove "permanent" or locked entries. We could probably do something better on spitfire, such as limiting the flush to kernel TLB entries or even doing range comparisons. However that probably isn't worth it since those chips are old and the TLB only had 64 entries. Reported-by: James Clarke <jrtc27@jrtc27.com> Tested-by: James Clarke <jrtc27@jrtc27.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-28 00:04:54 +08:00
2: mov %o3, %g1
mov %o5, %g3
mov 0, %o0 /* ARG0: CPU lists unimplemented */
mov 0, %o1 /* ARG1: CPU lists unimplemented */
mov 0, %o2 /* ARG2: mmu context == nucleus */
mov HV_MMU_ALL, %o3 /* ARG3: flags */
mov HV_FAST_MMU_DEMAP_CTX, %o5
ta HV_FAST_TRAP
mov %g1, %o3
brz,pt %o0, 5b
mov %g3, %o5
mov HV_FAST_MMU_DEMAP_CTX, %g6
ba,pt %xcc, 1b
clr %g5
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* These just get rescheduled to PIL vectors. */
.globl xcall_call_function
xcall_call_function:
wr %g0, (1 << PIL_SMP_CALL_FUNC), %set_softint
retry
sparc64: Convert to generic helpers for IPI function calls. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-07-18 14:44:50 +08:00
.globl xcall_call_function_single
xcall_call_function_single:
wr %g0, (1 << PIL_SMP_CALL_FUNC_SNGL), %set_softint
retry
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
.globl xcall_receive_signal
xcall_receive_signal:
wr %g0, (1 << PIL_SMP_RECEIVE_SIGNAL), %set_softint
retry
.globl xcall_capture
xcall_capture:
wr %g0, (1 << PIL_SMP_CAPTURE), %set_softint
retry
[SPARC64]: Fix new context version SMP handling. Don't piggy back the SMP receive signal code to do the context version change handling. Instead allocate another fixed PIL number for this asynchronous cross-call. We can't use smp_call_function() because this thing is invoked with interrupts disabled and a few spinlocks held. Also, fix smp_call_function_mask() to count "cpus" correctly. There is no guarentee that the local cpu is in the mask yet that is exactly what this code was assuming. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-03-07 14:50:44 +08:00
.globl xcall_new_mmu_context_version
xcall_new_mmu_context_version:
wr %g0, (1 << PIL_SMP_CTX_NEW_VERSION), %set_softint
retry
sparc: Add kgdb support. Current limitations: 1) On SMP single stepping has some fundamental issues, shared with other sw single-step architectures such as mips and arm. 2) On 32-bit sparc we don't support SMP kgdb yet. That requires some reworking of the IPI mechanisms and infrastructure on that platform. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-29 17:38:50 +08:00
#ifdef CONFIG_KGDB
.globl xcall_kgdb_capture
xcall_kgdb_capture:
sparc64: Reschedule KGDB capture to a software interrupt. Otherwise it might interrupt switch_to() midstream and use half-cooked register window state. Reported-by: Chris Torek <chris.torek@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-19 14:51:57 +08:00
wr %g0, (1 << PIL_KGDB_CAPTURE), %set_softint
retry
sparc: Add kgdb support. Current limitations: 1) On SMP single stepping has some fundamental issues, shared with other sw single-step architectures such as mips and arm. 2) On 32-bit sparc we don't support SMP kgdb yet. That requires some reworking of the IPI mechanisms and infrastructure on that platform. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-04-29 17:38:50 +08:00
#endif
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
#endif /* CONFIG_SMP */
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
sparc64: Handle extremely large kernel TLB range flushes more gracefully. When the vmalloc area gets fragmented, and because the firmware mapping area sits between where modules live and the vmalloc area, we can sometimes receive requests for enormous kernel TLB range flushes. When this happens the cpu just spins flushing billions of pages and this triggers the NMI watchdog and other problems. We took care of this on the TSB side by doing a linear scan of the table once we pass a certain threshold. Do something similar for the TLB flush, however we are limited by the TLB flush facilities provided by the different chip variants. First of all we use an (mostly arbitrary) cut-off of 256K which is about 32 pages. This can be tuned in the future. The huge range code path for each chip works as follows: 1) On spitfire we flush all non-locked TLB entries using diagnostic acceses. 2) On cheetah we use the "flush all" TLB flush. 3) On sun4v/hypervisor we do a TLB context flush on context 0, which unlike previous chips does not remove "permanent" or locked entries. We could probably do something better on spitfire, such as limiting the flush to kernel TLB entries or even doing range comparisons. However that probably isn't worth it since those chips are old and the TLB only had 64 entries. Reported-by: James Clarke <jrtc27@jrtc27.com> Tested-by: James Clarke <jrtc27@jrtc27.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-28 00:04:54 +08:00
.globl cheetah_patch_cachetlbops
cheetah_patch_cachetlbops:
save %sp, -128, %sp
sethi %hi(__flush_tlb_mm), %o0
or %o0, %lo(__flush_tlb_mm), %o0
sethi %hi(__cheetah_flush_tlb_mm), %o1
or %o1, %lo(__cheetah_flush_tlb_mm), %o1
call tlb_patch_one
mov 19, %o2
sethi %hi(__flush_tlb_page), %o0
or %o0, %lo(__flush_tlb_page), %o0
sethi %hi(__cheetah_flush_tlb_page), %o1
or %o1, %lo(__cheetah_flush_tlb_page), %o1
call tlb_patch_one
mov 22, %o2
sethi %hi(__flush_tlb_pending), %o0
or %o0, %lo(__flush_tlb_pending), %o0
sethi %hi(__cheetah_flush_tlb_pending), %o1
or %o1, %lo(__cheetah_flush_tlb_pending), %o1
call tlb_patch_one
mov 27, %o2
sethi %hi(__flush_tlb_kernel_range), %o0
or %o0, %lo(__flush_tlb_kernel_range), %o0
sethi %hi(__cheetah_flush_tlb_kernel_range), %o1
or %o1, %lo(__cheetah_flush_tlb_kernel_range), %o1
call tlb_patch_one
mov 31, %o2
#ifdef DCACHE_ALIASING_POSSIBLE
sethi %hi(__flush_dcache_page), %o0
or %o0, %lo(__flush_dcache_page), %o0
sethi %hi(__cheetah_flush_dcache_page), %o1
or %o1, %lo(__cheetah_flush_dcache_page), %o1
call tlb_patch_one
mov 11, %o2
#endif /* DCACHE_ALIASING_POSSIBLE */
#ifdef CONFIG_SMP
sethi %hi(xcall_flush_tlb_kernel_range), %o0
or %o0, %lo(xcall_flush_tlb_kernel_range), %o0
sethi %hi(__cheetah_xcall_flush_tlb_kernel_range), %o1
or %o1, %lo(__cheetah_xcall_flush_tlb_kernel_range), %o1
call tlb_patch_one
mov 44, %o2
#endif /* CONFIG_SMP */
ret
restore
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
.globl hypervisor_patch_cachetlbops
hypervisor_patch_cachetlbops:
save %sp, -128, %sp
sethi %hi(__flush_tlb_mm), %o0
or %o0, %lo(__flush_tlb_mm), %o0
sethi %hi(__hypervisor_flush_tlb_mm), %o1
or %o1, %lo(__hypervisor_flush_tlb_mm), %o1
call tlb_patch_one
sparc64: Fix illegal relative branches in hypervisor patched TLB code. When we copy code over to patch another piece of code, we can only use PC-relative branches that target code within that piece of code. Such PC-relative branches cannot be made to external symbols because the patch moves the location of the code and thus modifies the relative address of external symbols. Use an absolute jmpl to fix this problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-26 07:23:26 +08:00
mov 19, %o2
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
sparc64: Fix race in TLB batch processing. As reported by Dave Kleikamp, when we emit cross calls to do batched TLB flush processing we have a race because we do not synchronize on the sibling cpus completing the cross call. So meanwhile the TLB batch can be reset (tb->tlb_nr set to zero, etc.) and either flushes are missed or flushes will flush the wrong addresses. Fix this by using generic infrastructure to synchonize on the completion of the cross call. This first required getting the flush_tlb_pending() call out from switch_to() which operates with locks held and interrupts disabled. The problem is that smp_call_function_many() cannot be invoked with IRQs disabled and this is explicitly checked for with WARN_ON_ONCE(). We get the batch processing outside of locked IRQ disabled sections by using some ideas from the powerpc port. Namely, we only batch inside of arch_{enter,leave}_lazy_mmu_mode() calls. If we're not in such a region, we flush TLBs synchronously. 1) Get rid of xcall_flush_tlb_pending and per-cpu type implementations. 2) Do TLB batch cross calls instead via: smp_call_function_many() tlb_pending_func() __flush_tlb_pending() 3) Batch only in lazy mmu sequences: a) Add 'active' member to struct tlb_batch b) Define __HAVE_ARCH_ENTER_LAZY_MMU_MODE c) Set 'active' in arch_enter_lazy_mmu_mode() d) Run batch and clear 'active' in arch_leave_lazy_mmu_mode() e) Check 'active' in tlb_batch_add_one() and do a synchronous flush if it's clear. 4) Add infrastructure for synchronous TLB page flushes. a) Implement __flush_tlb_page and per-cpu variants, patch as needed. b) Likewise for xcall_flush_tlb_page. c) Implement smp_flush_tlb_page() to invoke the cross-call. d) Wire up global_flush_tlb_page() to the right routine based upon CONFIG_SMP 5) It turns out that singleton batches are very common, 2 out of every 3 batch flushes have only a single entry in them. The batch flush waiting is very expensive, both because of the poll on sibling cpu completeion, as well as because passing the tlb batch pointer to the sibling cpus invokes a shared memory dereference. Therefore, in flush_tlb_pending(), if there is only one entry in the batch perform a completely asynchronous global_flush_tlb_page() instead. Reported-by: Dave Kleikamp <dave.kleikamp@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Dave Kleikamp <dave.kleikamp@oracle.com>
2013-04-20 05:26:26 +08:00
sethi %hi(__flush_tlb_page), %o0
or %o0, %lo(__flush_tlb_page), %o0
sethi %hi(__hypervisor_flush_tlb_page), %o1
or %o1, %lo(__hypervisor_flush_tlb_page), %o1
call tlb_patch_one
sparc64: Fix illegal relative branches in hypervisor patched TLB code. When we copy code over to patch another piece of code, we can only use PC-relative branches that target code within that piece of code. Such PC-relative branches cannot be made to external symbols because the patch moves the location of the code and thus modifies the relative address of external symbols. Use an absolute jmpl to fix this problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-26 07:23:26 +08:00
mov 22, %o2
sparc64: Fix race in TLB batch processing. As reported by Dave Kleikamp, when we emit cross calls to do batched TLB flush processing we have a race because we do not synchronize on the sibling cpus completing the cross call. So meanwhile the TLB batch can be reset (tb->tlb_nr set to zero, etc.) and either flushes are missed or flushes will flush the wrong addresses. Fix this by using generic infrastructure to synchonize on the completion of the cross call. This first required getting the flush_tlb_pending() call out from switch_to() which operates with locks held and interrupts disabled. The problem is that smp_call_function_many() cannot be invoked with IRQs disabled and this is explicitly checked for with WARN_ON_ONCE(). We get the batch processing outside of locked IRQ disabled sections by using some ideas from the powerpc port. Namely, we only batch inside of arch_{enter,leave}_lazy_mmu_mode() calls. If we're not in such a region, we flush TLBs synchronously. 1) Get rid of xcall_flush_tlb_pending and per-cpu type implementations. 2) Do TLB batch cross calls instead via: smp_call_function_many() tlb_pending_func() __flush_tlb_pending() 3) Batch only in lazy mmu sequences: a) Add 'active' member to struct tlb_batch b) Define __HAVE_ARCH_ENTER_LAZY_MMU_MODE c) Set 'active' in arch_enter_lazy_mmu_mode() d) Run batch and clear 'active' in arch_leave_lazy_mmu_mode() e) Check 'active' in tlb_batch_add_one() and do a synchronous flush if it's clear. 4) Add infrastructure for synchronous TLB page flushes. a) Implement __flush_tlb_page and per-cpu variants, patch as needed. b) Likewise for xcall_flush_tlb_page. c) Implement smp_flush_tlb_page() to invoke the cross-call. d) Wire up global_flush_tlb_page() to the right routine based upon CONFIG_SMP 5) It turns out that singleton batches are very common, 2 out of every 3 batch flushes have only a single entry in them. The batch flush waiting is very expensive, both because of the poll on sibling cpu completeion, as well as because passing the tlb batch pointer to the sibling cpus invokes a shared memory dereference. Therefore, in flush_tlb_pending(), if there is only one entry in the batch perform a completely asynchronous global_flush_tlb_page() instead. Reported-by: Dave Kleikamp <dave.kleikamp@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Dave Kleikamp <dave.kleikamp@oracle.com>
2013-04-20 05:26:26 +08:00
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
sethi %hi(__flush_tlb_pending), %o0
or %o0, %lo(__flush_tlb_pending), %o0
sethi %hi(__hypervisor_flush_tlb_pending), %o1
or %o1, %lo(__hypervisor_flush_tlb_pending), %o1
call tlb_patch_one
sparc64: Fix illegal relative branches in hypervisor patched TLB code. When we copy code over to patch another piece of code, we can only use PC-relative branches that target code within that piece of code. Such PC-relative branches cannot be made to external symbols because the patch moves the location of the code and thus modifies the relative address of external symbols. Use an absolute jmpl to fix this problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-26 07:23:26 +08:00
mov 27, %o2
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
sethi %hi(__flush_tlb_kernel_range), %o0
or %o0, %lo(__flush_tlb_kernel_range), %o0
sethi %hi(__hypervisor_flush_tlb_kernel_range), %o1
or %o1, %lo(__hypervisor_flush_tlb_kernel_range), %o1
call tlb_patch_one
sparc64: Handle extremely large kernel TLB range flushes more gracefully. When the vmalloc area gets fragmented, and because the firmware mapping area sits between where modules live and the vmalloc area, we can sometimes receive requests for enormous kernel TLB range flushes. When this happens the cpu just spins flushing billions of pages and this triggers the NMI watchdog and other problems. We took care of this on the TSB side by doing a linear scan of the table once we pass a certain threshold. Do something similar for the TLB flush, however we are limited by the TLB flush facilities provided by the different chip variants. First of all we use an (mostly arbitrary) cut-off of 256K which is about 32 pages. This can be tuned in the future. The huge range code path for each chip works as follows: 1) On spitfire we flush all non-locked TLB entries using diagnostic acceses. 2) On cheetah we use the "flush all" TLB flush. 3) On sun4v/hypervisor we do a TLB context flush on context 0, which unlike previous chips does not remove "permanent" or locked entries. We could probably do something better on spitfire, such as limiting the flush to kernel TLB entries or even doing range comparisons. However that probably isn't worth it since those chips are old and the TLB only had 64 entries. Reported-by: James Clarke <jrtc27@jrtc27.com> Tested-by: James Clarke <jrtc27@jrtc27.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-28 00:04:54 +08:00
mov 31, %o2
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
#ifdef DCACHE_ALIASING_POSSIBLE
sethi %hi(__flush_dcache_page), %o0
or %o0, %lo(__flush_dcache_page), %o0
sethi %hi(__hypervisor_flush_dcache_page), %o1
or %o1, %lo(__hypervisor_flush_dcache_page), %o1
call tlb_patch_one
mov 2, %o2
#endif /* DCACHE_ALIASING_POSSIBLE */
#ifdef CONFIG_SMP
sethi %hi(xcall_flush_tlb_mm), %o0
or %o0, %lo(xcall_flush_tlb_mm), %o0
sethi %hi(__hypervisor_xcall_flush_tlb_mm), %o1
or %o1, %lo(__hypervisor_xcall_flush_tlb_mm), %o1
call tlb_patch_one
sparc64: Fix illegal relative branches in hypervisor patched TLB cross-call code. Just like the non-cross-call TLB flush handlers, the cross-call ones need to avoid doing PC-relative branches outside of their code blocks. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-27 01:20:14 +08:00
mov 24, %o2
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
sparc64: Fix race in TLB batch processing. As reported by Dave Kleikamp, when we emit cross calls to do batched TLB flush processing we have a race because we do not synchronize on the sibling cpus completing the cross call. So meanwhile the TLB batch can be reset (tb->tlb_nr set to zero, etc.) and either flushes are missed or flushes will flush the wrong addresses. Fix this by using generic infrastructure to synchonize on the completion of the cross call. This first required getting the flush_tlb_pending() call out from switch_to() which operates with locks held and interrupts disabled. The problem is that smp_call_function_many() cannot be invoked with IRQs disabled and this is explicitly checked for with WARN_ON_ONCE(). We get the batch processing outside of locked IRQ disabled sections by using some ideas from the powerpc port. Namely, we only batch inside of arch_{enter,leave}_lazy_mmu_mode() calls. If we're not in such a region, we flush TLBs synchronously. 1) Get rid of xcall_flush_tlb_pending and per-cpu type implementations. 2) Do TLB batch cross calls instead via: smp_call_function_many() tlb_pending_func() __flush_tlb_pending() 3) Batch only in lazy mmu sequences: a) Add 'active' member to struct tlb_batch b) Define __HAVE_ARCH_ENTER_LAZY_MMU_MODE c) Set 'active' in arch_enter_lazy_mmu_mode() d) Run batch and clear 'active' in arch_leave_lazy_mmu_mode() e) Check 'active' in tlb_batch_add_one() and do a synchronous flush if it's clear. 4) Add infrastructure for synchronous TLB page flushes. a) Implement __flush_tlb_page and per-cpu variants, patch as needed. b) Likewise for xcall_flush_tlb_page. c) Implement smp_flush_tlb_page() to invoke the cross-call. d) Wire up global_flush_tlb_page() to the right routine based upon CONFIG_SMP 5) It turns out that singleton batches are very common, 2 out of every 3 batch flushes have only a single entry in them. The batch flush waiting is very expensive, both because of the poll on sibling cpu completeion, as well as because passing the tlb batch pointer to the sibling cpus invokes a shared memory dereference. Therefore, in flush_tlb_pending(), if there is only one entry in the batch perform a completely asynchronous global_flush_tlb_page() instead. Reported-by: Dave Kleikamp <dave.kleikamp@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Acked-by: Dave Kleikamp <dave.kleikamp@oracle.com>
2013-04-20 05:26:26 +08:00
sethi %hi(xcall_flush_tlb_page), %o0
or %o0, %lo(xcall_flush_tlb_page), %o0
sethi %hi(__hypervisor_xcall_flush_tlb_page), %o1
or %o1, %lo(__hypervisor_xcall_flush_tlb_page), %o1
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
call tlb_patch_one
sparc64: Fix illegal relative branches in hypervisor patched TLB cross-call code. Just like the non-cross-call TLB flush handlers, the cross-call ones need to avoid doing PC-relative branches outside of their code blocks. Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-27 01:20:14 +08:00
mov 20, %o2
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
sethi %hi(xcall_flush_tlb_kernel_range), %o0
or %o0, %lo(xcall_flush_tlb_kernel_range), %o0
sethi %hi(__hypervisor_xcall_flush_tlb_kernel_range), %o1
or %o1, %lo(__hypervisor_xcall_flush_tlb_kernel_range), %o1
call tlb_patch_one
sparc64: Handle extremely large kernel TLB range flushes more gracefully. When the vmalloc area gets fragmented, and because the firmware mapping area sits between where modules live and the vmalloc area, we can sometimes receive requests for enormous kernel TLB range flushes. When this happens the cpu just spins flushing billions of pages and this triggers the NMI watchdog and other problems. We took care of this on the TSB side by doing a linear scan of the table once we pass a certain threshold. Do something similar for the TLB flush, however we are limited by the TLB flush facilities provided by the different chip variants. First of all we use an (mostly arbitrary) cut-off of 256K which is about 32 pages. This can be tuned in the future. The huge range code path for each chip works as follows: 1) On spitfire we flush all non-locked TLB entries using diagnostic acceses. 2) On cheetah we use the "flush all" TLB flush. 3) On sun4v/hypervisor we do a TLB context flush on context 0, which unlike previous chips does not remove "permanent" or locked entries. We could probably do something better on spitfire, such as limiting the flush to kernel TLB entries or even doing range comparisons. However that probably isn't worth it since those chips are old and the TLB only had 64 entries. Reported-by: James Clarke <jrtc27@jrtc27.com> Tested-by: James Clarke <jrtc27@jrtc27.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-28 00:04:54 +08:00
mov 44, %o2
[SPARC64]: SUN4V hypervisor TLB flush support code. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-04 19:08:37 +08:00
#endif /* CONFIG_SMP */
ret
restore