* [PULL 01/22] accel: Replace target_ulong in tlb_*()
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
@ 2023-06-26 15:39 ` Richard Henderson
2023-06-26 15:39 ` [PULL 02/22] accel/tcg/translate-all.c: Widen pc and cs_base Richard Henderson
` (21 subsequent siblings)
22 siblings, 0 replies; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 15:39 UTC (permalink / raw)
To: qemu-devel; +Cc: Anton Johansson
From: Anton Johansson <anjo@rev.ng>
Replaces target_ulong with vaddr for guest virtual addresses in tlb_*()
functions and auxilliary structs.
Signed-off-by: Anton Johansson <anjo@rev.ng>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230621135633.1649-2-anjo@rev.ng>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
include/exec/cpu-defs.h | 4 +-
include/exec/exec-all.h | 79 ++++++++--------
include/qemu/plugin-memory.h | 2 +-
accel/stubs/tcg-stub.c | 2 +-
accel/tcg/cputlb.c | 177 +++++++++++++++++------------------
accel/tcg/tb-maint.c | 2 +-
6 files changed, 131 insertions(+), 135 deletions(-)
diff --git a/include/exec/cpu-defs.h b/include/exec/cpu-defs.h
index 4cb77c8dec..e6a079402e 100644
--- a/include/exec/cpu-defs.h
+++ b/include/exec/cpu-defs.h
@@ -147,8 +147,8 @@ typedef struct CPUTLBDesc {
* we must flush the entire tlb. The region is matched if
* (addr & large_page_mask) == large_page_addr.
*/
- target_ulong large_page_addr;
- target_ulong large_page_mask;
+ vaddr large_page_addr;
+ vaddr large_page_mask;
/* host time (in ns) at the beginning of the time window */
int64_t window_begin_ns;
/* maximum number of entries observed in the window */
diff --git a/include/exec/exec-all.h b/include/exec/exec-all.h
index 698943d58f..f5508e242b 100644
--- a/include/exec/exec-all.h
+++ b/include/exec/exec-all.h
@@ -94,7 +94,7 @@ void tlb_destroy(CPUState *cpu);
* Flush one page from the TLB of the specified CPU, for all
* MMU indexes.
*/
-void tlb_flush_page(CPUState *cpu, target_ulong addr);
+void tlb_flush_page(CPUState *cpu, vaddr addr);
/**
* tlb_flush_page_all_cpus:
* @cpu: src CPU of the flush
@@ -103,7 +103,7 @@ void tlb_flush_page(CPUState *cpu, target_ulong addr);
* Flush one page from the TLB of the specified CPU, for all
* MMU indexes.
*/
-void tlb_flush_page_all_cpus(CPUState *src, target_ulong addr);
+void tlb_flush_page_all_cpus(CPUState *src, vaddr addr);
/**
* tlb_flush_page_all_cpus_synced:
* @cpu: src CPU of the flush
@@ -115,7 +115,7 @@ void tlb_flush_page_all_cpus(CPUState *src, target_ulong addr);
* the source vCPUs safe work is complete. This will depend on when
* the guests translation ends the TB.
*/
-void tlb_flush_page_all_cpus_synced(CPUState *src, target_ulong addr);
+void tlb_flush_page_all_cpus_synced(CPUState *src, vaddr addr);
/**
* tlb_flush:
* @cpu: CPU whose TLB should be flushed
@@ -150,7 +150,7 @@ void tlb_flush_all_cpus_synced(CPUState *src_cpu);
* Flush one page from the TLB of the specified CPU, for the specified
* MMU indexes.
*/
-void tlb_flush_page_by_mmuidx(CPUState *cpu, target_ulong addr,
+void tlb_flush_page_by_mmuidx(CPUState *cpu, vaddr addr,
uint16_t idxmap);
/**
* tlb_flush_page_by_mmuidx_all_cpus:
@@ -161,7 +161,7 @@ void tlb_flush_page_by_mmuidx(CPUState *cpu, target_ulong addr,
* Flush one page from the TLB of all CPUs, for the specified
* MMU indexes.
*/
-void tlb_flush_page_by_mmuidx_all_cpus(CPUState *cpu, target_ulong addr,
+void tlb_flush_page_by_mmuidx_all_cpus(CPUState *cpu, vaddr addr,
uint16_t idxmap);
/**
* tlb_flush_page_by_mmuidx_all_cpus_synced:
@@ -175,7 +175,7 @@ void tlb_flush_page_by_mmuidx_all_cpus(CPUState *cpu, target_ulong addr,
* complete once the source vCPUs safe work is complete. This will
* depend on when the guests translation ends the TB.
*/
-void tlb_flush_page_by_mmuidx_all_cpus_synced(CPUState *cpu, target_ulong addr,
+void tlb_flush_page_by_mmuidx_all_cpus_synced(CPUState *cpu, vaddr addr,
uint16_t idxmap);
/**
* tlb_flush_by_mmuidx:
@@ -218,14 +218,14 @@ void tlb_flush_by_mmuidx_all_cpus_synced(CPUState *cpu, uint16_t idxmap);
*
* Similar to tlb_flush_page_mask, but with a bitmap of indexes.
*/
-void tlb_flush_page_bits_by_mmuidx(CPUState *cpu, target_ulong addr,
+void tlb_flush_page_bits_by_mmuidx(CPUState *cpu, vaddr addr,
uint16_t idxmap, unsigned bits);
/* Similarly, with broadcast and syncing. */
-void tlb_flush_page_bits_by_mmuidx_all_cpus(CPUState *cpu, target_ulong addr,
+void tlb_flush_page_bits_by_mmuidx_all_cpus(CPUState *cpu, vaddr addr,
uint16_t idxmap, unsigned bits);
void tlb_flush_page_bits_by_mmuidx_all_cpus_synced
- (CPUState *cpu, target_ulong addr, uint16_t idxmap, unsigned bits);
+ (CPUState *cpu, vaddr addr, uint16_t idxmap, unsigned bits);
/**
* tlb_flush_range_by_mmuidx
@@ -238,17 +238,17 @@ void tlb_flush_page_bits_by_mmuidx_all_cpus_synced
* For each mmuidx in @idxmap, flush all pages within [@addr,@addr+@len),
* comparing only the low @bits worth of each virtual page.
*/
-void tlb_flush_range_by_mmuidx(CPUState *cpu, target_ulong addr,
- target_ulong len, uint16_t idxmap,
+void tlb_flush_range_by_mmuidx(CPUState *cpu, vaddr addr,
+ vaddr len, uint16_t idxmap,
unsigned bits);
/* Similarly, with broadcast and syncing. */
-void tlb_flush_range_by_mmuidx_all_cpus(CPUState *cpu, target_ulong addr,
- target_ulong len, uint16_t idxmap,
+void tlb_flush_range_by_mmuidx_all_cpus(CPUState *cpu, vaddr addr,
+ vaddr len, uint16_t idxmap,
unsigned bits);
void tlb_flush_range_by_mmuidx_all_cpus_synced(CPUState *cpu,
- target_ulong addr,
- target_ulong len,
+ vaddr addr,
+ vaddr len,
uint16_t idxmap,
unsigned bits);
@@ -256,7 +256,7 @@ void tlb_flush_range_by_mmuidx_all_cpus_synced(CPUState *cpu,
* tlb_set_page_full:
* @cpu: CPU context
* @mmu_idx: mmu index of the tlb to modify
- * @vaddr: virtual address of the entry to add
+ * @addr: virtual address of the entry to add
* @full: the details of the tlb entry
*
* Add an entry to @cpu tlb index @mmu_idx. All of the fields of
@@ -271,13 +271,13 @@ void tlb_flush_range_by_mmuidx_all_cpus_synced(CPUState *cpu,
* single TARGET_PAGE_SIZE region is mapped; @full->lg_page_size is only
* used by tlb_flush_page.
*/
-void tlb_set_page_full(CPUState *cpu, int mmu_idx, target_ulong vaddr,
+void tlb_set_page_full(CPUState *cpu, int mmu_idx, vaddr addr,
CPUTLBEntryFull *full);
/**
* tlb_set_page_with_attrs:
* @cpu: CPU to add this TLB entry for
- * @vaddr: virtual address of page to add entry for
+ * @addr: virtual address of page to add entry for
* @paddr: physical address of the page
* @attrs: memory transaction attributes
* @prot: access permissions (PAGE_READ/PAGE_WRITE/PAGE_EXEC bits)
@@ -285,7 +285,7 @@ void tlb_set_page_full(CPUState *cpu, int mmu_idx, target_ulong vaddr,
* @size: size of the page in bytes
*
* Add an entry to this CPU's TLB (a mapping from virtual address
- * @vaddr to physical address @paddr) with the specified memory
+ * @addr to physical address @paddr) with the specified memory
* transaction attributes. This is generally called by the target CPU
* specific code after it has been called through the tlb_fill()
* entry point and performed a successful page table walk to find
@@ -296,18 +296,18 @@ void tlb_set_page_full(CPUState *cpu, int mmu_idx, target_ulong vaddr,
* single TARGET_PAGE_SIZE region is mapped; the supplied @size is only
* used by tlb_flush_page.
*/
-void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr,
+void tlb_set_page_with_attrs(CPUState *cpu, vaddr addr,
hwaddr paddr, MemTxAttrs attrs,
- int prot, int mmu_idx, target_ulong size);
+ int prot, int mmu_idx, vaddr size);
/* tlb_set_page:
*
* This function is equivalent to calling tlb_set_page_with_attrs()
* with an @attrs argument of MEMTXATTRS_UNSPECIFIED. It's provided
* as a convenience for CPUs which don't use memory transaction attributes.
*/
-void tlb_set_page(CPUState *cpu, target_ulong vaddr,
+void tlb_set_page(CPUState *cpu, vaddr addr,
hwaddr paddr, int prot,
- int mmu_idx, target_ulong size);
+ int mmu_idx, vaddr size);
#else
static inline void tlb_init(CPUState *cpu)
{
@@ -315,14 +315,13 @@ static inline void tlb_init(CPUState *cpu)
static inline void tlb_destroy(CPUState *cpu)
{
}
-static inline void tlb_flush_page(CPUState *cpu, target_ulong addr)
+static inline void tlb_flush_page(CPUState *cpu, vaddr addr)
{
}
-static inline void tlb_flush_page_all_cpus(CPUState *src, target_ulong addr)
+static inline void tlb_flush_page_all_cpus(CPUState *src, vaddr addr)
{
}
-static inline void tlb_flush_page_all_cpus_synced(CPUState *src,
- target_ulong addr)
+static inline void tlb_flush_page_all_cpus_synced(CPUState *src, vaddr addr)
{
}
static inline void tlb_flush(CPUState *cpu)
@@ -335,7 +334,7 @@ static inline void tlb_flush_all_cpus_synced(CPUState *src_cpu)
{
}
static inline void tlb_flush_page_by_mmuidx(CPUState *cpu,
- target_ulong addr, uint16_t idxmap)
+ vaddr addr, uint16_t idxmap)
{
}
@@ -343,12 +342,12 @@ static inline void tlb_flush_by_mmuidx(CPUState *cpu, uint16_t idxmap)
{
}
static inline void tlb_flush_page_by_mmuidx_all_cpus(CPUState *cpu,
- target_ulong addr,
+ vaddr addr,
uint16_t idxmap)
{
}
static inline void tlb_flush_page_by_mmuidx_all_cpus_synced(CPUState *cpu,
- target_ulong addr,
+ vaddr addr,
uint16_t idxmap)
{
}
@@ -361,37 +360,37 @@ static inline void tlb_flush_by_mmuidx_all_cpus_synced(CPUState *cpu,
{
}
static inline void tlb_flush_page_bits_by_mmuidx(CPUState *cpu,
- target_ulong addr,
+ vaddr addr,
uint16_t idxmap,
unsigned bits)
{
}
static inline void tlb_flush_page_bits_by_mmuidx_all_cpus(CPUState *cpu,
- target_ulong addr,
+ vaddr addr,
uint16_t idxmap,
unsigned bits)
{
}
static inline void
-tlb_flush_page_bits_by_mmuidx_all_cpus_synced(CPUState *cpu, target_ulong addr,
+tlb_flush_page_bits_by_mmuidx_all_cpus_synced(CPUState *cpu, vaddr addr,
uint16_t idxmap, unsigned bits)
{
}
-static inline void tlb_flush_range_by_mmuidx(CPUState *cpu, target_ulong addr,
- target_ulong len, uint16_t idxmap,
+static inline void tlb_flush_range_by_mmuidx(CPUState *cpu, vaddr addr,
+ vaddr len, uint16_t idxmap,
unsigned bits)
{
}
static inline void tlb_flush_range_by_mmuidx_all_cpus(CPUState *cpu,
- target_ulong addr,
- target_ulong len,
+ vaddr addr,
+ vaddr len,
uint16_t idxmap,
unsigned bits)
{
}
static inline void tlb_flush_range_by_mmuidx_all_cpus_synced(CPUState *cpu,
- target_ulong addr,
- target_long len,
+ vaddr addr,
+ vaddr len,
uint16_t idxmap,
unsigned bits)
{
@@ -663,7 +662,7 @@ static inline void mmap_lock(void) {}
static inline void mmap_unlock(void) {}
void tlb_reset_dirty(CPUState *cpu, ram_addr_t start1, ram_addr_t length);
-void tlb_set_dirty(CPUState *cpu, target_ulong vaddr);
+void tlb_set_dirty(CPUState *cpu, vaddr addr);
MemoryRegionSection *
address_space_translate_for_iotlb(CPUState *cpu, int asidx, hwaddr addr,
diff --git a/include/qemu/plugin-memory.h b/include/qemu/plugin-memory.h
index 6fd539022a..43165f2452 100644
--- a/include/qemu/plugin-memory.h
+++ b/include/qemu/plugin-memory.h
@@ -37,7 +37,7 @@ struct qemu_plugin_hwaddr {
* It would only fail if not called from an instrumented memory access
* which would be an abuse of the API.
*/
-bool tlb_plugin_lookup(CPUState *cpu, target_ulong addr, int mmu_idx,
+bool tlb_plugin_lookup(CPUState *cpu, vaddr addr, int mmu_idx,
bool is_store, struct qemu_plugin_hwaddr *data);
#endif /* PLUGIN_MEMORY_H */
diff --git a/accel/stubs/tcg-stub.c b/accel/stubs/tcg-stub.c
index 813695b402..0998e601ad 100644
--- a/accel/stubs/tcg-stub.c
+++ b/accel/stubs/tcg-stub.c
@@ -18,7 +18,7 @@ void tb_flush(CPUState *cpu)
{
}
-void tlb_set_dirty(CPUState *cpu, target_ulong vaddr)
+void tlb_set_dirty(CPUState *cpu, vaddr vaddr)
{
}
diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c
index 14ce97c33b..5caeccb52d 100644
--- a/accel/tcg/cputlb.c
+++ b/accel/tcg/cputlb.c
@@ -427,7 +427,7 @@ void tlb_flush_all_cpus_synced(CPUState *src_cpu)
}
static bool tlb_hit_page_mask_anyprot(CPUTLBEntry *tlb_entry,
- target_ulong page, target_ulong mask)
+ vaddr page, vaddr mask)
{
page &= mask;
mask &= TARGET_PAGE_MASK | TLB_INVALID_MASK;
@@ -437,8 +437,7 @@ static bool tlb_hit_page_mask_anyprot(CPUTLBEntry *tlb_entry,
page == (tlb_entry->addr_code & mask));
}
-static inline bool tlb_hit_page_anyprot(CPUTLBEntry *tlb_entry,
- target_ulong page)
+static inline bool tlb_hit_page_anyprot(CPUTLBEntry *tlb_entry, vaddr page)
{
return tlb_hit_page_mask_anyprot(tlb_entry, page, -1);
}
@@ -454,8 +453,8 @@ static inline bool tlb_entry_is_empty(const CPUTLBEntry *te)
/* Called with tlb_c.lock held */
static bool tlb_flush_entry_mask_locked(CPUTLBEntry *tlb_entry,
- target_ulong page,
- target_ulong mask)
+ vaddr page,
+ vaddr mask)
{
if (tlb_hit_page_mask_anyprot(tlb_entry, page, mask)) {
memset(tlb_entry, -1, sizeof(*tlb_entry));
@@ -464,16 +463,15 @@ static bool tlb_flush_entry_mask_locked(CPUTLBEntry *tlb_entry,
return false;
}
-static inline bool tlb_flush_entry_locked(CPUTLBEntry *tlb_entry,
- target_ulong page)
+static inline bool tlb_flush_entry_locked(CPUTLBEntry *tlb_entry, vaddr page)
{
return tlb_flush_entry_mask_locked(tlb_entry, page, -1);
}
/* Called with tlb_c.lock held */
static void tlb_flush_vtlb_page_mask_locked(CPUArchState *env, int mmu_idx,
- target_ulong page,
- target_ulong mask)
+ vaddr page,
+ vaddr mask)
{
CPUTLBDesc *d = &env_tlb(env)->d[mmu_idx];
int k;
@@ -487,21 +485,20 @@ static void tlb_flush_vtlb_page_mask_locked(CPUArchState *env, int mmu_idx,
}
static inline void tlb_flush_vtlb_page_locked(CPUArchState *env, int mmu_idx,
- target_ulong page)
+ vaddr page)
{
tlb_flush_vtlb_page_mask_locked(env, mmu_idx, page, -1);
}
-static void tlb_flush_page_locked(CPUArchState *env, int midx,
- target_ulong page)
+static void tlb_flush_page_locked(CPUArchState *env, int midx, vaddr page)
{
- target_ulong lp_addr = env_tlb(env)->d[midx].large_page_addr;
- target_ulong lp_mask = env_tlb(env)->d[midx].large_page_mask;
+ vaddr lp_addr = env_tlb(env)->d[midx].large_page_addr;
+ vaddr lp_mask = env_tlb(env)->d[midx].large_page_mask;
/* Check if we need to flush due to large pages. */
if ((page & lp_mask) == lp_addr) {
- tlb_debug("forcing full flush midx %d ("
- TARGET_FMT_lx "/" TARGET_FMT_lx ")\n",
+ tlb_debug("forcing full flush midx %d (%"
+ VADDR_PRIx "/%" VADDR_PRIx ")\n",
midx, lp_addr, lp_mask);
tlb_flush_one_mmuidx_locked(env, midx, get_clock_realtime());
} else {
@@ -522,7 +519,7 @@ static void tlb_flush_page_locked(CPUArchState *env, int midx,
* at @addr from the tlbs indicated by @idxmap from @cpu.
*/
static void tlb_flush_page_by_mmuidx_async_0(CPUState *cpu,
- target_ulong addr,
+ vaddr addr,
uint16_t idxmap)
{
CPUArchState *env = cpu->env_ptr;
@@ -530,7 +527,7 @@ static void tlb_flush_page_by_mmuidx_async_0(CPUState *cpu,
assert_cpu_is_self(cpu);
- tlb_debug("page addr:" TARGET_FMT_lx " mmu_map:0x%x\n", addr, idxmap);
+ tlb_debug("page addr: %" VADDR_PRIx " mmu_map:0x%x\n", addr, idxmap);
qemu_spin_lock(&env_tlb(env)->c.lock);
for (mmu_idx = 0; mmu_idx < NB_MMU_MODES; mmu_idx++) {
@@ -561,15 +558,15 @@ static void tlb_flush_page_by_mmuidx_async_0(CPUState *cpu,
static void tlb_flush_page_by_mmuidx_async_1(CPUState *cpu,
run_on_cpu_data data)
{
- target_ulong addr_and_idxmap = (target_ulong) data.target_ptr;
- target_ulong addr = addr_and_idxmap & TARGET_PAGE_MASK;
+ vaddr addr_and_idxmap = data.target_ptr;
+ vaddr addr = addr_and_idxmap & TARGET_PAGE_MASK;
uint16_t idxmap = addr_and_idxmap & ~TARGET_PAGE_MASK;
tlb_flush_page_by_mmuidx_async_0(cpu, addr, idxmap);
}
typedef struct {
- target_ulong addr;
+ vaddr addr;
uint16_t idxmap;
} TLBFlushPageByMMUIdxData;
@@ -592,9 +589,9 @@ static void tlb_flush_page_by_mmuidx_async_2(CPUState *cpu,
g_free(d);
}
-void tlb_flush_page_by_mmuidx(CPUState *cpu, target_ulong addr, uint16_t idxmap)
+void tlb_flush_page_by_mmuidx(CPUState *cpu, vaddr addr, uint16_t idxmap)
{
- tlb_debug("addr: "TARGET_FMT_lx" mmu_idx:%" PRIx16 "\n", addr, idxmap);
+ tlb_debug("addr: %" VADDR_PRIx " mmu_idx:%" PRIx16 "\n", addr, idxmap);
/* This should already be page aligned */
addr &= TARGET_PAGE_MASK;
@@ -620,15 +617,15 @@ void tlb_flush_page_by_mmuidx(CPUState *cpu, target_ulong addr, uint16_t idxmap)
}
}
-void tlb_flush_page(CPUState *cpu, target_ulong addr)
+void tlb_flush_page(CPUState *cpu, vaddr addr)
{
tlb_flush_page_by_mmuidx(cpu, addr, ALL_MMUIDX_BITS);
}
-void tlb_flush_page_by_mmuidx_all_cpus(CPUState *src_cpu, target_ulong addr,
+void tlb_flush_page_by_mmuidx_all_cpus(CPUState *src_cpu, vaddr addr,
uint16_t idxmap)
{
- tlb_debug("addr: "TARGET_FMT_lx" mmu_idx:%"PRIx16"\n", addr, idxmap);
+ tlb_debug("addr: %" VADDR_PRIx " mmu_idx:%"PRIx16"\n", addr, idxmap);
/* This should already be page aligned */
addr &= TARGET_PAGE_MASK;
@@ -660,16 +657,16 @@ void tlb_flush_page_by_mmuidx_all_cpus(CPUState *src_cpu, target_ulong addr,
tlb_flush_page_by_mmuidx_async_0(src_cpu, addr, idxmap);
}
-void tlb_flush_page_all_cpus(CPUState *src, target_ulong addr)
+void tlb_flush_page_all_cpus(CPUState *src, vaddr addr)
{
tlb_flush_page_by_mmuidx_all_cpus(src, addr, ALL_MMUIDX_BITS);
}
void tlb_flush_page_by_mmuidx_all_cpus_synced(CPUState *src_cpu,
- target_ulong addr,
+ vaddr addr,
uint16_t idxmap)
{
- tlb_debug("addr: "TARGET_FMT_lx" mmu_idx:%"PRIx16"\n", addr, idxmap);
+ tlb_debug("addr: %" VADDR_PRIx " mmu_idx:%"PRIx16"\n", addr, idxmap);
/* This should already be page aligned */
addr &= TARGET_PAGE_MASK;
@@ -706,18 +703,18 @@ void tlb_flush_page_by_mmuidx_all_cpus_synced(CPUState *src_cpu,
}
}
-void tlb_flush_page_all_cpus_synced(CPUState *src, target_ulong addr)
+void tlb_flush_page_all_cpus_synced(CPUState *src, vaddr addr)
{
tlb_flush_page_by_mmuidx_all_cpus_synced(src, addr, ALL_MMUIDX_BITS);
}
static void tlb_flush_range_locked(CPUArchState *env, int midx,
- target_ulong addr, target_ulong len,
+ vaddr addr, vaddr len,
unsigned bits)
{
CPUTLBDesc *d = &env_tlb(env)->d[midx];
CPUTLBDescFast *f = &env_tlb(env)->f[midx];
- target_ulong mask = MAKE_64BIT_MASK(0, bits);
+ vaddr mask = MAKE_64BIT_MASK(0, bits);
/*
* If @bits is smaller than the tlb size, there may be multiple entries
@@ -731,7 +728,7 @@ static void tlb_flush_range_locked(CPUArchState *env, int midx,
*/
if (mask < f->mask || len > f->mask) {
tlb_debug("forcing full flush midx %d ("
- TARGET_FMT_lx "/" TARGET_FMT_lx "+" TARGET_FMT_lx ")\n",
+ "%" VADDR_PRIx "/%" VADDR_PRIx "+%" VADDR_PRIx ")\n",
midx, addr, mask, len);
tlb_flush_one_mmuidx_locked(env, midx, get_clock_realtime());
return;
@@ -744,14 +741,14 @@ static void tlb_flush_range_locked(CPUArchState *env, int midx,
*/
if (((addr + len - 1) & d->large_page_mask) == d->large_page_addr) {
tlb_debug("forcing full flush midx %d ("
- TARGET_FMT_lx "/" TARGET_FMT_lx ")\n",
+ "%" VADDR_PRIx "/%" VADDR_PRIx ")\n",
midx, d->large_page_addr, d->large_page_mask);
tlb_flush_one_mmuidx_locked(env, midx, get_clock_realtime());
return;
}
- for (target_ulong i = 0; i < len; i += TARGET_PAGE_SIZE) {
- target_ulong page = addr + i;
+ for (vaddr i = 0; i < len; i += TARGET_PAGE_SIZE) {
+ vaddr page = addr + i;
CPUTLBEntry *entry = tlb_entry(env, midx, page);
if (tlb_flush_entry_mask_locked(entry, page, mask)) {
@@ -762,8 +759,8 @@ static void tlb_flush_range_locked(CPUArchState *env, int midx,
}
typedef struct {
- target_ulong addr;
- target_ulong len;
+ vaddr addr;
+ vaddr len;
uint16_t idxmap;
uint16_t bits;
} TLBFlushRangeData;
@@ -776,7 +773,7 @@ static void tlb_flush_range_by_mmuidx_async_0(CPUState *cpu,
assert_cpu_is_self(cpu);
- tlb_debug("range:" TARGET_FMT_lx "/%u+" TARGET_FMT_lx " mmu_map:0x%x\n",
+ tlb_debug("range: %" VADDR_PRIx "/%u+%" VADDR_PRIx " mmu_map:0x%x\n",
d.addr, d.bits, d.len, d.idxmap);
qemu_spin_lock(&env_tlb(env)->c.lock);
@@ -801,7 +798,7 @@ static void tlb_flush_range_by_mmuidx_async_0(CPUState *cpu,
* overlap the flushed pages, which includes the previous.
*/
d.addr -= TARGET_PAGE_SIZE;
- for (target_ulong i = 0, n = d.len / TARGET_PAGE_SIZE + 1; i < n; i++) {
+ for (vaddr i = 0, n = d.len / TARGET_PAGE_SIZE + 1; i < n; i++) {
tb_jmp_cache_clear_page(cpu, d.addr);
d.addr += TARGET_PAGE_SIZE;
}
@@ -815,8 +812,8 @@ static void tlb_flush_range_by_mmuidx_async_1(CPUState *cpu,
g_free(d);
}
-void tlb_flush_range_by_mmuidx(CPUState *cpu, target_ulong addr,
- target_ulong len, uint16_t idxmap,
+void tlb_flush_range_by_mmuidx(CPUState *cpu, vaddr addr,
+ vaddr len, uint16_t idxmap,
unsigned bits)
{
TLBFlushRangeData d;
@@ -851,14 +848,14 @@ void tlb_flush_range_by_mmuidx(CPUState *cpu, target_ulong addr,
}
}
-void tlb_flush_page_bits_by_mmuidx(CPUState *cpu, target_ulong addr,
+void tlb_flush_page_bits_by_mmuidx(CPUState *cpu, vaddr addr,
uint16_t idxmap, unsigned bits)
{
tlb_flush_range_by_mmuidx(cpu, addr, TARGET_PAGE_SIZE, idxmap, bits);
}
void tlb_flush_range_by_mmuidx_all_cpus(CPUState *src_cpu,
- target_ulong addr, target_ulong len,
+ vaddr addr, vaddr len,
uint16_t idxmap, unsigned bits)
{
TLBFlushRangeData d;
@@ -898,16 +895,16 @@ void tlb_flush_range_by_mmuidx_all_cpus(CPUState *src_cpu,
}
void tlb_flush_page_bits_by_mmuidx_all_cpus(CPUState *src_cpu,
- target_ulong addr,
- uint16_t idxmap, unsigned bits)
+ vaddr addr, uint16_t idxmap,
+ unsigned bits)
{
tlb_flush_range_by_mmuidx_all_cpus(src_cpu, addr, TARGET_PAGE_SIZE,
idxmap, bits);
}
void tlb_flush_range_by_mmuidx_all_cpus_synced(CPUState *src_cpu,
- target_ulong addr,
- target_ulong len,
+ vaddr addr,
+ vaddr len,
uint16_t idxmap,
unsigned bits)
{
@@ -949,7 +946,7 @@ void tlb_flush_range_by_mmuidx_all_cpus_synced(CPUState *src_cpu,
}
void tlb_flush_page_bits_by_mmuidx_all_cpus_synced(CPUState *src_cpu,
- target_ulong addr,
+ vaddr addr,
uint16_t idxmap,
unsigned bits)
{
@@ -1055,32 +1052,32 @@ void tlb_reset_dirty(CPUState *cpu, ram_addr_t start1, ram_addr_t length)
/* Called with tlb_c.lock held */
static inline void tlb_set_dirty1_locked(CPUTLBEntry *tlb_entry,
- target_ulong vaddr)
+ vaddr addr)
{
- if (tlb_entry->addr_write == (vaddr | TLB_NOTDIRTY)) {
- tlb_entry->addr_write = vaddr;
+ if (tlb_entry->addr_write == (addr | TLB_NOTDIRTY)) {
+ tlb_entry->addr_write = addr;
}
}
/* update the TLB corresponding to virtual page vaddr
so that it is no longer dirty */
-void tlb_set_dirty(CPUState *cpu, target_ulong vaddr)
+void tlb_set_dirty(CPUState *cpu, vaddr addr)
{
CPUArchState *env = cpu->env_ptr;
int mmu_idx;
assert_cpu_is_self(cpu);
- vaddr &= TARGET_PAGE_MASK;
+ addr &= TARGET_PAGE_MASK;
qemu_spin_lock(&env_tlb(env)->c.lock);
for (mmu_idx = 0; mmu_idx < NB_MMU_MODES; mmu_idx++) {
- tlb_set_dirty1_locked(tlb_entry(env, mmu_idx, vaddr), vaddr);
+ tlb_set_dirty1_locked(tlb_entry(env, mmu_idx, addr), addr);
}
for (mmu_idx = 0; mmu_idx < NB_MMU_MODES; mmu_idx++) {
int k;
for (k = 0; k < CPU_VTLB_SIZE; k++) {
- tlb_set_dirty1_locked(&env_tlb(env)->d[mmu_idx].vtable[k], vaddr);
+ tlb_set_dirty1_locked(&env_tlb(env)->d[mmu_idx].vtable[k], addr);
}
}
qemu_spin_unlock(&env_tlb(env)->c.lock);
@@ -1089,20 +1086,20 @@ void tlb_set_dirty(CPUState *cpu, target_ulong vaddr)
/* Our TLB does not support large pages, so remember the area covered by
large pages and trigger a full TLB flush if these are invalidated. */
static void tlb_add_large_page(CPUArchState *env, int mmu_idx,
- target_ulong vaddr, target_ulong size)
+ vaddr addr, uint64_t size)
{
- target_ulong lp_addr = env_tlb(env)->d[mmu_idx].large_page_addr;
- target_ulong lp_mask = ~(size - 1);
+ vaddr lp_addr = env_tlb(env)->d[mmu_idx].large_page_addr;
+ vaddr lp_mask = ~(size - 1);
- if (lp_addr == (target_ulong)-1) {
+ if (lp_addr == (vaddr)-1) {
/* No previous large page. */
- lp_addr = vaddr;
+ lp_addr = addr;
} else {
/* Extend the existing region to include the new page.
This is a compromise between unnecessary flushes and
the cost of maintaining a full variable size TLB. */
lp_mask &= env_tlb(env)->d[mmu_idx].large_page_mask;
- while (((lp_addr ^ vaddr) & lp_mask) != 0) {
+ while (((lp_addr ^ addr) & lp_mask) != 0) {
lp_mask <<= 1;
}
}
@@ -1119,19 +1116,19 @@ static void tlb_add_large_page(CPUArchState *env, int mmu_idx,
* critical section.
*/
void tlb_set_page_full(CPUState *cpu, int mmu_idx,
- target_ulong vaddr, CPUTLBEntryFull *full)
+ vaddr addr, CPUTLBEntryFull *full)
{
CPUArchState *env = cpu->env_ptr;
CPUTLB *tlb = env_tlb(env);
CPUTLBDesc *desc = &tlb->d[mmu_idx];
MemoryRegionSection *section;
unsigned int index;
- target_ulong address;
- target_ulong write_address;
+ vaddr address;
+ vaddr write_address;
uintptr_t addend;
CPUTLBEntry *te, tn;
hwaddr iotlb, xlat, sz, paddr_page;
- target_ulong vaddr_page;
+ vaddr addr_page;
int asidx, wp_flags, prot;
bool is_ram, is_romd;
@@ -1141,9 +1138,9 @@ void tlb_set_page_full(CPUState *cpu, int mmu_idx,
sz = TARGET_PAGE_SIZE;
} else {
sz = (hwaddr)1 << full->lg_page_size;
- tlb_add_large_page(env, mmu_idx, vaddr, sz);
+ tlb_add_large_page(env, mmu_idx, addr, sz);
}
- vaddr_page = vaddr & TARGET_PAGE_MASK;
+ addr_page = addr & TARGET_PAGE_MASK;
paddr_page = full->phys_addr & TARGET_PAGE_MASK;
prot = full->prot;
@@ -1152,11 +1149,11 @@ void tlb_set_page_full(CPUState *cpu, int mmu_idx,
&xlat, &sz, full->attrs, &prot);
assert(sz >= TARGET_PAGE_SIZE);
- tlb_debug("vaddr=" TARGET_FMT_lx " paddr=0x" HWADDR_FMT_plx
+ tlb_debug("vaddr=%" VADDR_PRIx " paddr=0x" HWADDR_FMT_plx
" prot=%x idx=%d\n",
- vaddr, full->phys_addr, prot, mmu_idx);
+ addr, full->phys_addr, prot, mmu_idx);
- address = vaddr_page;
+ address = addr_page;
if (full->lg_page_size < TARGET_PAGE_BITS) {
/* Repeat the MMU check and TLB fill on every access. */
address |= TLB_INVALID_MASK;
@@ -1204,11 +1201,11 @@ void tlb_set_page_full(CPUState *cpu, int mmu_idx,
}
}
- wp_flags = cpu_watchpoint_address_matches(cpu, vaddr_page,
+ wp_flags = cpu_watchpoint_address_matches(cpu, addr_page,
TARGET_PAGE_SIZE);
- index = tlb_index(env, mmu_idx, vaddr_page);
- te = tlb_entry(env, mmu_idx, vaddr_page);
+ index = tlb_index(env, mmu_idx, addr_page);
+ te = tlb_entry(env, mmu_idx, addr_page);
/*
* Hold the TLB lock for the rest of the function. We could acquire/release
@@ -1223,13 +1220,13 @@ void tlb_set_page_full(CPUState *cpu, int mmu_idx,
tlb->c.dirty |= 1 << mmu_idx;
/* Make sure there's no cached translation for the new page. */
- tlb_flush_vtlb_page_locked(env, mmu_idx, vaddr_page);
+ tlb_flush_vtlb_page_locked(env, mmu_idx, addr_page);
/*
* Only evict the old entry to the victim tlb if it's for a
* different page; otherwise just overwrite the stale data.
*/
- if (!tlb_hit_page_anyprot(te, vaddr_page) && !tlb_entry_is_empty(te)) {
+ if (!tlb_hit_page_anyprot(te, addr_page) && !tlb_entry_is_empty(te)) {
unsigned vidx = desc->vindex++ % CPU_VTLB_SIZE;
CPUTLBEntry *tv = &desc->vtable[vidx];
@@ -1253,11 +1250,11 @@ void tlb_set_page_full(CPUState *cpu, int mmu_idx,
* vaddr we add back in io_readx()/io_writex()/get_page_addr_code().
*/
desc->fulltlb[index] = *full;
- desc->fulltlb[index].xlat_section = iotlb - vaddr_page;
+ desc->fulltlb[index].xlat_section = iotlb - addr_page;
desc->fulltlb[index].phys_addr = paddr_page;
/* Now calculate the new entry */
- tn.addend = addend - vaddr_page;
+ tn.addend = addend - addr_page;
if (prot & PAGE_READ) {
tn.addr_read = address;
if (wp_flags & BP_MEM_READ) {
@@ -1289,9 +1286,9 @@ void tlb_set_page_full(CPUState *cpu, int mmu_idx,
qemu_spin_unlock(&tlb->c.lock);
}
-void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr,
+void tlb_set_page_with_attrs(CPUState *cpu, vaddr addr,
hwaddr paddr, MemTxAttrs attrs, int prot,
- int mmu_idx, target_ulong size)
+ int mmu_idx, uint64_t size)
{
CPUTLBEntryFull full = {
.phys_addr = paddr,
@@ -1301,14 +1298,14 @@ void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr,
};
assert(is_power_of_2(size));
- tlb_set_page_full(cpu, mmu_idx, vaddr, &full);
+ tlb_set_page_full(cpu, mmu_idx, addr, &full);
}
-void tlb_set_page(CPUState *cpu, target_ulong vaddr,
+void tlb_set_page(CPUState *cpu, vaddr addr,
hwaddr paddr, int prot,
- int mmu_idx, target_ulong size)
+ int mmu_idx, uint64_t size)
{
- tlb_set_page_with_attrs(cpu, vaddr, paddr, MEMTXATTRS_UNSPECIFIED,
+ tlb_set_page_with_attrs(cpu, addr, paddr, MEMTXATTRS_UNSPECIFIED,
prot, mmu_idx, size);
}
@@ -1317,7 +1314,7 @@ void tlb_set_page(CPUState *cpu, target_ulong vaddr,
* caller's prior references to the TLB table (e.g. CPUTLBEntry pointers) must
* be discarded and looked up again (e.g. via tlb_entry()).
*/
-static void tlb_fill(CPUState *cpu, target_ulong addr, int size,
+static void tlb_fill(CPUState *cpu, vaddr addr, int size,
MMUAccessType access_type, int mmu_idx, uintptr_t retaddr)
{
bool ok;
@@ -1357,7 +1354,7 @@ static inline void cpu_transaction_failed(CPUState *cpu, hwaddr physaddr,
}
static uint64_t io_readx(CPUArchState *env, CPUTLBEntryFull *full,
- int mmu_idx, target_ulong addr, uintptr_t retaddr,
+ int mmu_idx, vaddr addr, uintptr_t retaddr,
MMUAccessType access_type, MemOp op)
{
CPUState *cpu = env_cpu(env);
@@ -1407,7 +1404,7 @@ static void save_iotlb_data(CPUState *cs, MemoryRegionSection *section,
}
static void io_writex(CPUArchState *env, CPUTLBEntryFull *full,
- int mmu_idx, uint64_t val, target_ulong addr,
+ int mmu_idx, uint64_t val, vaddr addr,
uintptr_t retaddr, MemOp op)
{
CPUState *cpu = env_cpu(env);
@@ -1449,7 +1446,7 @@ static void io_writex(CPUArchState *env, CPUTLBEntryFull *full,
/* Return true if ADDR is present in the victim tlb, and has been copied
back to the main tlb. */
static bool victim_tlb_hit(CPUArchState *env, size_t mmu_idx, size_t index,
- MMUAccessType access_type, target_ulong page)
+ MMUAccessType access_type, vaddr page)
{
size_t vidx;
@@ -1691,13 +1688,13 @@ tb_page_addr_t get_page_addr_code_hostp(CPUArchState *env, target_ulong addr,
* from the same thread (which a mem callback will be) this is safe.
*/
-bool tlb_plugin_lookup(CPUState *cpu, target_ulong addr, int mmu_idx,
+bool tlb_plugin_lookup(CPUState *cpu, vaddr addr, int mmu_idx,
bool is_store, struct qemu_plugin_hwaddr *data)
{
CPUArchState *env = cpu->env_ptr;
CPUTLBEntry *tlbe = tlb_entry(env, mmu_idx, addr);
uintptr_t index = tlb_index(env, mmu_idx, addr);
- target_ulong tlb_addr = is_store ? tlb_addr_write(tlbe) : tlbe->addr_read;
+ vaddr tlb_addr = is_store ? tlb_addr_write(tlbe) : tlbe->addr_read;
if (likely(tlb_hit(tlb_addr, addr))) {
/* We must have an iotlb entry for MMIO */
diff --git a/accel/tcg/tb-maint.c b/accel/tcg/tb-maint.c
index 892eecda2d..3541419845 100644
--- a/accel/tcg/tb-maint.c
+++ b/accel/tcg/tb-maint.c
@@ -98,7 +98,7 @@ static void tb_remove_all(void)
/* Call with mmap_lock held. */
static void tb_record(TranslationBlock *tb, PageDesc *p1, PageDesc *p2)
{
- target_ulong addr;
+ vaddr addr;
int flags;
assert_memory_lock();
--
2.34.1
^ permalink raw reply related [flat|nested] 28+ messages in thread* [PULL 02/22] accel/tcg/translate-all.c: Widen pc and cs_base
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
2023-06-26 15:39 ` [PULL 01/22] accel: Replace target_ulong in tlb_*() Richard Henderson
@ 2023-06-26 15:39 ` Richard Henderson
2023-07-11 16:39 ` Peter Maydell
2023-06-26 15:39 ` [PULL 03/22] target: Widen pc/cs_base in cpu_get_tb_cpu_state Richard Henderson
` (20 subsequent siblings)
22 siblings, 1 reply; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 15:39 UTC (permalink / raw)
To: qemu-devel; +Cc: Anton Johansson
From: Anton Johansson <anjo@rev.ng>
Signed-off-by: Anton Johansson <anjo@rev.ng>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230621135633.1649-3-anjo@rev.ng>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
accel/tcg/internal.h | 6 +++---
accel/tcg/translate-all.c | 10 +++++-----
2 files changed, 8 insertions(+), 8 deletions(-)
diff --git a/accel/tcg/internal.h b/accel/tcg/internal.h
index 65380ccb42..91f308bdfa 100644
--- a/accel/tcg/internal.h
+++ b/accel/tcg/internal.h
@@ -42,8 +42,8 @@ void tb_invalidate_phys_range_fast(ram_addr_t ram_addr,
G_NORETURN void cpu_io_recompile(CPUState *cpu, uintptr_t retaddr);
#endif /* CONFIG_SOFTMMU */
-TranslationBlock *tb_gen_code(CPUState *cpu, target_ulong pc,
- target_ulong cs_base, uint32_t flags,
+TranslationBlock *tb_gen_code(CPUState *cpu, vaddr pc,
+ uint64_t cs_base, uint32_t flags,
int cflags);
void page_init(void);
void tb_htable_init(void);
@@ -55,7 +55,7 @@ void cpu_restore_state_from_tb(CPUState *cpu, TranslationBlock *tb,
uintptr_t host_pc);
/* Return the current PC from CPU, which may be cached in TB. */
-static inline target_ulong log_pc(CPUState *cpu, const TranslationBlock *tb)
+static inline vaddr log_pc(CPUState *cpu, const TranslationBlock *tb)
{
if (tb_cflags(tb) & CF_PCREL) {
return cpu->cc->get_pc(cpu);
diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c
index c4d081f5ad..03caf62459 100644
--- a/accel/tcg/translate-all.c
+++ b/accel/tcg/translate-all.c
@@ -274,7 +274,7 @@ void page_init(void)
* Return the size of the generated code, or negative on error.
*/
static int setjmp_gen_code(CPUArchState *env, TranslationBlock *tb,
- target_ulong pc, void *host_pc,
+ vaddr pc, void *host_pc,
int *max_insns, int64_t *ti)
{
int ret = sigsetjmp(tcg_ctx->jmp_trans, 0);
@@ -302,7 +302,7 @@ static int setjmp_gen_code(CPUArchState *env, TranslationBlock *tb,
/* Called with mmap_lock held for user mode emulation. */
TranslationBlock *tb_gen_code(CPUState *cpu,
- target_ulong pc, target_ulong cs_base,
+ vaddr pc, uint64_t cs_base,
uint32_t flags, int cflags)
{
CPUArchState *env = cpu->env_ptr;
@@ -634,10 +634,10 @@ void cpu_io_recompile(CPUState *cpu, uintptr_t retaddr)
cpu->cflags_next_tb = curr_cflags(cpu) | CF_MEMI_ONLY | CF_LAST_IO | n;
if (qemu_loglevel_mask(CPU_LOG_EXEC)) {
- target_ulong pc = log_pc(cpu, tb);
+ vaddr pc = log_pc(cpu, tb);
if (qemu_log_in_addr_range(pc)) {
- qemu_log("cpu_io_recompile: rewound execution of TB to "
- TARGET_FMT_lx "\n", pc);
+ qemu_log("cpu_io_recompile: rewound execution of TB to %"
+ VADDR_PRIx "\n", pc);
}
}
--
2.34.1
^ permalink raw reply related [flat|nested] 28+ messages in thread* Re: [PULL 02/22] accel/tcg/translate-all.c: Widen pc and cs_base
2023-06-26 15:39 ` [PULL 02/22] accel/tcg/translate-all.c: Widen pc and cs_base Richard Henderson
@ 2023-07-11 16:39 ` Peter Maydell
0 siblings, 0 replies; 28+ messages in thread
From: Peter Maydell @ 2023-07-11 16:39 UTC (permalink / raw)
To: Richard Henderson; +Cc: qemu-devel, Anton Johansson
On Mon, 26 Jun 2023 at 16:40, Richard Henderson
<richard.henderson@linaro.org> wrote:
>
> From: Anton Johansson <anjo@rev.ng>
>
> Signed-off-by: Anton Johansson <anjo@rev.ng>
> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
> Message-Id: <20230621135633.1649-3-anjo@rev.ng>
> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
> ---
> @@ -634,10 +634,10 @@ void cpu_io_recompile(CPUState *cpu, uintptr_t retaddr)
> cpu->cflags_next_tb = curr_cflags(cpu) | CF_MEMI_ONLY | CF_LAST_IO | n;
>
> if (qemu_loglevel_mask(CPU_LOG_EXEC)) {
> - target_ulong pc = log_pc(cpu, tb);
> + vaddr pc = log_pc(cpu, tb);
> if (qemu_log_in_addr_range(pc)) {
> - qemu_log("cpu_io_recompile: rewound execution of TB to "
> - TARGET_FMT_lx "\n", pc);
> + qemu_log("cpu_io_recompile: rewound execution of TB to %"
> + VADDR_PRIx "\n", pc);
TARGET_FMT_lx includes a width specifier so the value is
left-zero-padded to the appropriate width (depending
on the size of the PC on the target). VADDR_PRIx does
not zero-pad. Intentional change?
thanks
-- PMM
^ permalink raw reply [flat|nested] 28+ messages in thread
* [PULL 03/22] target: Widen pc/cs_base in cpu_get_tb_cpu_state
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
2023-06-26 15:39 ` [PULL 01/22] accel: Replace target_ulong in tlb_*() Richard Henderson
2023-06-26 15:39 ` [PULL 02/22] accel/tcg/translate-all.c: Widen pc and cs_base Richard Henderson
@ 2023-06-26 15:39 ` Richard Henderson
2023-06-26 15:39 ` [PULL 04/22] accel/tcg/cputlb.c: Widen CPUTLBEntry access functions Richard Henderson
` (19 subsequent siblings)
22 siblings, 0 replies; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 15:39 UTC (permalink / raw)
To: qemu-devel; +Cc: Anton Johansson
From: Anton Johansson <anjo@rev.ng>
Signed-off-by: Anton Johansson <anjo@rev.ng>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230621135633.1649-4-anjo@rev.ng>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
target/alpha/cpu.h | 4 ++--
target/arm/cpu.h | 4 ++--
target/avr/cpu.h | 4 ++--
target/cris/cpu.h | 4 ++--
target/hexagon/cpu.h | 4 ++--
target/hppa/cpu.h | 5 ++---
target/i386/cpu.h | 4 ++--
target/loongarch/cpu.h | 6 ++----
target/m68k/cpu.h | 4 ++--
target/microblaze/cpu.h | 4 ++--
target/mips/cpu.h | 4 ++--
target/nios2/cpu.h | 4 ++--
target/openrisc/cpu.h | 5 ++---
target/ppc/cpu.h | 8 ++++----
target/riscv/cpu.h | 4 ++--
target/rx/cpu.h | 4 ++--
target/s390x/cpu.h | 4 ++--
target/sh4/cpu.h | 4 ++--
target/sparc/cpu.h | 4 ++--
target/tricore/cpu.h | 4 ++--
target/xtensa/cpu.h | 4 ++--
accel/tcg/cpu-exec.c | 9 ++++++---
accel/tcg/translate-all.c | 3 ++-
target/arm/helper.c | 4 ++--
target/ppc/helper_regs.c | 4 ++--
target/riscv/cpu_helper.c | 4 ++--
26 files changed, 58 insertions(+), 58 deletions(-)
diff --git a/target/alpha/cpu.h b/target/alpha/cpu.h
index 5e67304d81..fcd20bfd3a 100644
--- a/target/alpha/cpu.h
+++ b/target/alpha/cpu.h
@@ -462,8 +462,8 @@ void alpha_cpu_do_transaction_failed(CPUState *cs, hwaddr physaddr,
MemTxResult response, uintptr_t retaddr);
#endif
-static inline void cpu_get_tb_cpu_state(CPUAlphaState *env, target_ulong *pc,
- target_ulong *cs_base, uint32_t *pflags)
+static inline void cpu_get_tb_cpu_state(CPUAlphaState *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *pflags)
{
*pc = env->pc;
*cs_base = 0;
diff --git a/target/arm/cpu.h b/target/arm/cpu.h
index 11c3850ad9..00e675f58f 100644
--- a/target/arm/cpu.h
+++ b/target/arm/cpu.h
@@ -3220,8 +3220,8 @@ static inline bool arm_cpu_bswap_data(CPUARMState *env)
}
#endif
-void cpu_get_tb_cpu_state(CPUARMState *env, target_ulong *pc,
- target_ulong *cs_base, uint32_t *flags);
+void cpu_get_tb_cpu_state(CPUARMState *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *flags);
enum {
QEMU_PSCI_CONDUIT_DISABLED = 0,
diff --git a/target/avr/cpu.h b/target/avr/cpu.h
index f19dd72926..7225174668 100644
--- a/target/avr/cpu.h
+++ b/target/avr/cpu.h
@@ -190,8 +190,8 @@ enum {
TB_FLAGS_SKIP = 2,
};
-static inline void cpu_get_tb_cpu_state(CPUAVRState *env, target_ulong *pc,
- target_ulong *cs_base, uint32_t *pflags)
+static inline void cpu_get_tb_cpu_state(CPUAVRState *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *pflags)
{
uint32_t flags = 0;
diff --git a/target/cris/cpu.h b/target/cris/cpu.h
index 71fa1f96e0..8e37c6e50d 100644
--- a/target/cris/cpu.h
+++ b/target/cris/cpu.h
@@ -266,8 +266,8 @@ static inline int cpu_mmu_index (CPUCRISState *env, bool ifetch)
#include "exec/cpu-all.h"
-static inline void cpu_get_tb_cpu_state(CPUCRISState *env, target_ulong *pc,
- target_ulong *cs_base, uint32_t *flags)
+static inline void cpu_get_tb_cpu_state(CPUCRISState *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *flags)
{
*pc = env->pc;
*cs_base = 0;
diff --git a/target/hexagon/cpu.h b/target/hexagon/cpu.h
index bfcb1057dd..daef5c3f00 100644
--- a/target/hexagon/cpu.h
+++ b/target/hexagon/cpu.h
@@ -153,8 +153,8 @@ struct ArchCPU {
FIELD(TB_FLAGS, IS_TIGHT_LOOP, 0, 1)
-static inline void cpu_get_tb_cpu_state(CPUHexagonState *env, target_ulong *pc,
- target_ulong *cs_base, uint32_t *flags)
+static inline void cpu_get_tb_cpu_state(CPUHexagonState *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *flags)
{
uint32_t hex_flags = 0;
*pc = env->gpr[HEX_REG_PC];
diff --git a/target/hppa/cpu.h b/target/hppa/cpu.h
index b595ef25a9..7373177b55 100644
--- a/target/hppa/cpu.h
+++ b/target/hppa/cpu.h
@@ -268,9 +268,8 @@ static inline target_ulong hppa_form_gva(CPUHPPAState *env, uint64_t spc,
#define TB_FLAG_PRIV_SHIFT 8
#define TB_FLAG_UNALIGN 0x400
-static inline void cpu_get_tb_cpu_state(CPUHPPAState *env, target_ulong *pc,
- target_ulong *cs_base,
- uint32_t *pflags)
+static inline void cpu_get_tb_cpu_state(CPUHPPAState *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *pflags)
{
uint32_t flags = env->psw_n * PSW_N;
diff --git a/target/i386/cpu.h b/target/i386/cpu.h
index cd047e0410..2c9b0d2ebc 100644
--- a/target/i386/cpu.h
+++ b/target/i386/cpu.h
@@ -2275,8 +2275,8 @@ static inline int cpu_mmu_index_kernel(CPUX86State *env)
#include "hw/i386/apic.h"
#endif
-static inline void cpu_get_tb_cpu_state(CPUX86State *env, target_ulong *pc,
- target_ulong *cs_base, uint32_t *flags)
+static inline void cpu_get_tb_cpu_state(CPUX86State *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *flags)
{
*cs_base = env->segs[R_CS].base;
*pc = *cs_base + env->eip;
diff --git a/target/loongarch/cpu.h b/target/loongarch/cpu.h
index b23f38c3d5..ed04027af1 100644
--- a/target/loongarch/cpu.h
+++ b/target/loongarch/cpu.h
@@ -427,10 +427,8 @@ static inline int cpu_mmu_index(CPULoongArchState *env, bool ifetch)
#define HW_FLAGS_EUEN_FPE 0x04
#define HW_FLAGS_EUEN_SXE 0x08
-static inline void cpu_get_tb_cpu_state(CPULoongArchState *env,
- target_ulong *pc,
- target_ulong *cs_base,
- uint32_t *flags)
+static inline void cpu_get_tb_cpu_state(CPULoongArchState *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *flags)
{
*pc = env->pc;
*cs_base = 0;
diff --git a/target/m68k/cpu.h b/target/m68k/cpu.h
index 048d5aae2b..cf70282717 100644
--- a/target/m68k/cpu.h
+++ b/target/m68k/cpu.h
@@ -601,8 +601,8 @@ void m68k_cpu_transaction_failed(CPUState *cs, hwaddr physaddr, vaddr addr,
#define TB_FLAGS_TRACE 16
#define TB_FLAGS_TRACE_BIT (1 << TB_FLAGS_TRACE)
-static inline void cpu_get_tb_cpu_state(CPUM68KState *env, target_ulong *pc,
- target_ulong *cs_base, uint32_t *flags)
+static inline void cpu_get_tb_cpu_state(CPUM68KState *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *flags)
{
*pc = env->pc;
*cs_base = 0;
diff --git a/target/microblaze/cpu.h b/target/microblaze/cpu.h
index 88324d0bc1..3525de144c 100644
--- a/target/microblaze/cpu.h
+++ b/target/microblaze/cpu.h
@@ -401,8 +401,8 @@ void mb_tcg_init(void);
/* Ensure there is no overlap between the two masks. */
QEMU_BUILD_BUG_ON(MSR_TB_MASK & IFLAGS_TB_MASK);
-static inline void cpu_get_tb_cpu_state(CPUMBState *env, target_ulong *pc,
- target_ulong *cs_base, uint32_t *flags)
+static inline void cpu_get_tb_cpu_state(CPUMBState *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *flags)
{
*pc = env->pc;
*flags = (env->iflags & IFLAGS_TB_MASK) | (env->msr & MSR_TB_MASK);
diff --git a/target/mips/cpu.h b/target/mips/cpu.h
index 142c55af47..a3bc646976 100644
--- a/target/mips/cpu.h
+++ b/target/mips/cpu.h
@@ -1313,8 +1313,8 @@ void itc_reconfigure(struct MIPSITUState *tag);
/* helper.c */
target_ulong exception_resume_pc(CPUMIPSState *env);
-static inline void cpu_get_tb_cpu_state(CPUMIPSState *env, target_ulong *pc,
- target_ulong *cs_base, uint32_t *flags)
+static inline void cpu_get_tb_cpu_state(CPUMIPSState *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *flags)
{
*pc = env->active_tc.PC;
*cs_base = 0;
diff --git a/target/nios2/cpu.h b/target/nios2/cpu.h
index 20042c4332..477a3161fd 100644
--- a/target/nios2/cpu.h
+++ b/target/nios2/cpu.h
@@ -302,8 +302,8 @@ FIELD(TBFLAGS, CRS0, 0, 1) /* Set if CRS == 0. */
FIELD(TBFLAGS, U, 1, 1) /* Overlaps CR_STATUS_U */
FIELD(TBFLAGS, R0_0, 2, 1) /* Set if R0 == 0. */
-static inline void cpu_get_tb_cpu_state(CPUNios2State *env, target_ulong *pc,
- target_ulong *cs_base, uint32_t *flags)
+static inline void cpu_get_tb_cpu_state(CPUNios2State *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *flags)
{
unsigned crs = FIELD_EX32(env->ctrl[CR_STATUS], CR_STATUS, CRS);
diff --git a/target/openrisc/cpu.h b/target/openrisc/cpu.h
index f16e8b3274..92c38f54c2 100644
--- a/target/openrisc/cpu.h
+++ b/target/openrisc/cpu.h
@@ -367,9 +367,8 @@ static inline void cpu_set_gpr(CPUOpenRISCState *env, int i, uint32_t val)
env->shadow_gpr[0][i] = val;
}
-static inline void cpu_get_tb_cpu_state(CPUOpenRISCState *env,
- target_ulong *pc,
- target_ulong *cs_base, uint32_t *flags)
+static inline void cpu_get_tb_cpu_state(CPUOpenRISCState *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *flags)
{
*pc = env->pc;
*cs_base = 0;
diff --git a/target/ppc/cpu.h b/target/ppc/cpu.h
index 4138a25801..94497aa115 100644
--- a/target/ppc/cpu.h
+++ b/target/ppc/cpu.h
@@ -2508,11 +2508,11 @@ void cpu_write_xer(CPUPPCState *env, target_ulong xer);
#define is_book3s_arch2x(ctx) (!!((ctx)->insns_flags & PPC_SEGMENT_64B))
#ifdef CONFIG_DEBUG_TCG
-void cpu_get_tb_cpu_state(CPUPPCState *env, target_ulong *pc,
- target_ulong *cs_base, uint32_t *flags);
+void cpu_get_tb_cpu_state(CPUPPCState *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *flags);
#else
-static inline void cpu_get_tb_cpu_state(CPUPPCState *env, target_ulong *pc,
- target_ulong *cs_base, uint32_t *flags)
+static inline void cpu_get_tb_cpu_state(CPUPPCState *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *flags)
{
*pc = env->nip;
*cs_base = 0;
diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
index e3e08d315f..7bff1d47f6 100644
--- a/target/riscv/cpu.h
+++ b/target/riscv/cpu.h
@@ -587,8 +587,8 @@ static inline uint32_t vext_get_vlmax(RISCVCPU *cpu, target_ulong vtype)
return cpu->cfg.vlen >> (sew + 3 - lmul);
}
-void cpu_get_tb_cpu_state(CPURISCVState *env, target_ulong *pc,
- target_ulong *cs_base, uint32_t *pflags);
+void cpu_get_tb_cpu_state(CPURISCVState *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *pflags);
void riscv_cpu_update_mask(CPURISCVState *env);
diff --git a/target/rx/cpu.h b/target/rx/cpu.h
index 555d230f24..7f03ffcfed 100644
--- a/target/rx/cpu.h
+++ b/target/rx/cpu.h
@@ -143,8 +143,8 @@ void rx_cpu_unpack_psw(CPURXState *env, uint32_t psw, int rte);
#define RX_CPU_IRQ 0
#define RX_CPU_FIR 1
-static inline void cpu_get_tb_cpu_state(CPURXState *env, target_ulong *pc,
- target_ulong *cs_base, uint32_t *flags)
+static inline void cpu_get_tb_cpu_state(CPURXState *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *flags)
{
*pc = env->pc;
*cs_base = 0;
diff --git a/target/s390x/cpu.h b/target/s390x/cpu.h
index f130c29f83..eb5b65b7d3 100644
--- a/target/s390x/cpu.h
+++ b/target/s390x/cpu.h
@@ -378,8 +378,8 @@ static inline int cpu_mmu_index(CPUS390XState *env, bool ifetch)
#endif
}
-static inline void cpu_get_tb_cpu_state(CPUS390XState* env, target_ulong *pc,
- target_ulong *cs_base, uint32_t *flags)
+static inline void cpu_get_tb_cpu_state(CPUS390XState *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *flags)
{
if (env->psw.addr & 1) {
/*
diff --git a/target/sh4/cpu.h b/target/sh4/cpu.h
index 02bfd612ea..1399d3840f 100644
--- a/target/sh4/cpu.h
+++ b/target/sh4/cpu.h
@@ -368,8 +368,8 @@ static inline void cpu_write_sr(CPUSH4State *env, target_ulong sr)
env->sr = sr & ~((1u << SR_M) | (1u << SR_Q) | (1u << SR_T));
}
-static inline void cpu_get_tb_cpu_state(CPUSH4State *env, target_ulong *pc,
- target_ulong *cs_base, uint32_t *flags)
+static inline void cpu_get_tb_cpu_state(CPUSH4State *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *flags)
{
*pc = env->pc;
/* For a gUSA region, notice the end of the region. */
diff --git a/target/sparc/cpu.h b/target/sparc/cpu.h
index 3d090e8278..95d2d0da71 100644
--- a/target/sparc/cpu.h
+++ b/target/sparc/cpu.h
@@ -762,8 +762,8 @@ trap_state* cpu_tsptr(CPUSPARCState* env);
#define TB_FLAG_HYPER (1 << 7)
#define TB_FLAG_ASI_SHIFT 24
-static inline void cpu_get_tb_cpu_state(CPUSPARCState *env, target_ulong *pc,
- target_ulong *cs_base, uint32_t *pflags)
+static inline void cpu_get_tb_cpu_state(CPUSPARCState *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *pflags)
{
uint32_t flags;
*pc = env->pc;
diff --git a/target/tricore/cpu.h b/target/tricore/cpu.h
index 257fcf3cee..a50b91cc36 100644
--- a/target/tricore/cpu.h
+++ b/target/tricore/cpu.h
@@ -384,8 +384,8 @@ FIELD(TB_FLAGS, PRIV, 0, 2)
void cpu_state_reset(CPUTriCoreState *s);
void tricore_tcg_init(void);
-static inline void cpu_get_tb_cpu_state(CPUTriCoreState *env, target_ulong *pc,
- target_ulong *cs_base, uint32_t *flags)
+static inline void cpu_get_tb_cpu_state(CPUTriCoreState *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *flags)
{
uint32_t new_flags = 0;
*pc = env->PC;
diff --git a/target/xtensa/cpu.h b/target/xtensa/cpu.h
index b7a54711a6..87fe992ba6 100644
--- a/target/xtensa/cpu.h
+++ b/target/xtensa/cpu.h
@@ -727,8 +727,8 @@ static inline int cpu_mmu_index(CPUXtensaState *env, bool ifetch)
#include "exec/cpu-all.h"
-static inline void cpu_get_tb_cpu_state(CPUXtensaState *env, target_ulong *pc,
- target_ulong *cs_base, uint32_t *flags)
+static inline void cpu_get_tb_cpu_state(CPUXtensaState *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *flags)
{
*pc = env->pc;
*cs_base = 0;
diff --git a/accel/tcg/cpu-exec.c b/accel/tcg/cpu-exec.c
index 179847b294..4d952a6cc2 100644
--- a/accel/tcg/cpu-exec.c
+++ b/accel/tcg/cpu-exec.c
@@ -408,7 +408,8 @@ const void *HELPER(lookup_tb_ptr)(CPUArchState *env)
{
CPUState *cpu = env_cpu(env);
TranslationBlock *tb;
- target_ulong cs_base, pc;
+ vaddr pc;
+ uint64_t cs_base;
uint32_t flags, cflags;
cpu_get_tb_cpu_state(env, &pc, &cs_base, &flags);
@@ -529,7 +530,8 @@ void cpu_exec_step_atomic(CPUState *cpu)
{
CPUArchState *env = cpu->env_ptr;
TranslationBlock *tb;
- target_ulong cs_base, pc;
+ vaddr pc;
+ uint64_t cs_base;
uint32_t flags, cflags;
int tb_exit;
@@ -942,7 +944,8 @@ cpu_exec_loop(CPUState *cpu, SyncClocks *sc)
while (!cpu_handle_interrupt(cpu, &last_tb)) {
TranslationBlock *tb;
- target_ulong cs_base, pc;
+ vaddr pc;
+ uint64_t cs_base;
uint32_t flags, cflags;
cpu_get_tb_cpu_state(cpu->env_ptr, &pc, &cs_base, &flags);
diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c
index 03caf62459..03c49baf1c 100644
--- a/accel/tcg/translate-all.c
+++ b/accel/tcg/translate-all.c
@@ -580,7 +580,8 @@ void tb_check_watchpoint(CPUState *cpu, uintptr_t retaddr)
/* The exception probably happened in a helper. The CPU state should
have been saved before calling it. Fetch the PC from there. */
CPUArchState *env = cpu->env_ptr;
- target_ulong pc, cs_base;
+ vaddr pc;
+ uint64_t cs_base;
tb_page_addr_t addr;
uint32_t flags;
diff --git a/target/arm/helper.c b/target/arm/helper.c
index 323cadd3c8..d08c058e42 100644
--- a/target/arm/helper.c
+++ b/target/arm/helper.c
@@ -11945,8 +11945,8 @@ static bool mve_no_pred(CPUARMState *env)
return true;
}
-void cpu_get_tb_cpu_state(CPUARMState *env, target_ulong *pc,
- target_ulong *cs_base, uint32_t *pflags)
+void cpu_get_tb_cpu_state(CPUARMState *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *pflags)
{
CPUARMTBFlags flags;
diff --git a/target/ppc/helper_regs.c b/target/ppc/helper_regs.c
index e27f4a75a4..f380342d4d 100644
--- a/target/ppc/helper_regs.c
+++ b/target/ppc/helper_regs.c
@@ -218,8 +218,8 @@ void hreg_update_pmu_hflags(CPUPPCState *env)
}
#ifdef CONFIG_DEBUG_TCG
-void cpu_get_tb_cpu_state(CPUPPCState *env, target_ulong *pc,
- target_ulong *cs_base, uint32_t *flags)
+void cpu_get_tb_cpu_state(CPUPPCState *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *flags)
{
uint32_t hflags_current = env->hflags;
uint32_t hflags_rebuilt;
diff --git a/target/riscv/cpu_helper.c b/target/riscv/cpu_helper.c
index 90cef9856d..a944f25694 100644
--- a/target/riscv/cpu_helper.c
+++ b/target/riscv/cpu_helper.c
@@ -61,8 +61,8 @@ int riscv_cpu_mmu_index(CPURISCVState *env, bool ifetch)
#endif
}
-void cpu_get_tb_cpu_state(CPURISCVState *env, target_ulong *pc,
- target_ulong *cs_base, uint32_t *pflags)
+void cpu_get_tb_cpu_state(CPURISCVState *env, vaddr *pc,
+ uint64_t *cs_base, uint32_t *pflags)
{
CPUState *cs = env_cpu(env);
RISCVCPU *cpu = RISCV_CPU(cs);
--
2.34.1
^ permalink raw reply related [flat|nested] 28+ messages in thread* [PULL 04/22] accel/tcg/cputlb.c: Widen CPUTLBEntry access functions
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
` (2 preceding siblings ...)
2023-06-26 15:39 ` [PULL 03/22] target: Widen pc/cs_base in cpu_get_tb_cpu_state Richard Henderson
@ 2023-06-26 15:39 ` Richard Henderson
2023-06-26 15:39 ` [PULL 05/22] accel/tcg/cputlb.c: Widen addr in MMULookupPageData Richard Henderson
` (18 subsequent siblings)
22 siblings, 0 replies; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 15:39 UTC (permalink / raw)
To: qemu-devel; +Cc: Anton Johansson
From: Anton Johansson <anjo@rev.ng>
Signed-off-by: Anton Johansson <anjo@rev.ng>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230621135633.1649-5-anjo@rev.ng>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
include/exec/cpu_ldst.h | 10 +++++-----
accel/tcg/cputlb.c | 8 ++++----
2 files changed, 9 insertions(+), 9 deletions(-)
diff --git a/include/exec/cpu_ldst.h b/include/exec/cpu_ldst.h
index 896f305ff3..645476f0e5 100644
--- a/include/exec/cpu_ldst.h
+++ b/include/exec/cpu_ldst.h
@@ -328,8 +328,8 @@ static inline void clear_helper_retaddr(void)
#include "tcg/oversized-guest.h"
-static inline target_ulong tlb_read_idx(const CPUTLBEntry *entry,
- MMUAccessType access_type)
+static inline uint64_t tlb_read_idx(const CPUTLBEntry *entry,
+ MMUAccessType access_type)
{
/* Do not rearrange the CPUTLBEntry structure members. */
QEMU_BUILD_BUG_ON(offsetof(CPUTLBEntry, addr_read) !=
@@ -355,14 +355,14 @@ static inline target_ulong tlb_read_idx(const CPUTLBEntry *entry,
#endif
}
-static inline target_ulong tlb_addr_write(const CPUTLBEntry *entry)
+static inline uint64_t tlb_addr_write(const CPUTLBEntry *entry)
{
return tlb_read_idx(entry, MMU_DATA_STORE);
}
/* Find the TLB index corresponding to the mmu_idx + address pair. */
static inline uintptr_t tlb_index(CPUArchState *env, uintptr_t mmu_idx,
- target_ulong addr)
+ vaddr addr)
{
uintptr_t size_mask = env_tlb(env)->f[mmu_idx].mask >> CPU_TLB_ENTRY_BITS;
@@ -371,7 +371,7 @@ static inline uintptr_t tlb_index(CPUArchState *env, uintptr_t mmu_idx,
/* Find the TLB entry corresponding to the mmu_idx + address pair. */
static inline CPUTLBEntry *tlb_entry(CPUArchState *env, uintptr_t mmu_idx,
- target_ulong addr)
+ vaddr addr)
{
return &env_tlb(env)->f[mmu_idx].table[tlb_index(env, mmu_idx, addr)];
}
diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c
index 5caeccb52d..ac990a1526 100644
--- a/accel/tcg/cputlb.c
+++ b/accel/tcg/cputlb.c
@@ -1453,7 +1453,7 @@ static bool victim_tlb_hit(CPUArchState *env, size_t mmu_idx, size_t index,
assert_cpu_is_self(env_cpu(env));
for (vidx = 0; vidx < CPU_VTLB_SIZE; ++vidx) {
CPUTLBEntry *vtlb = &env_tlb(env)->d[mmu_idx].vtable[vidx];
- target_ulong cmp = tlb_read_idx(vtlb, access_type);
+ uint64_t cmp = tlb_read_idx(vtlb, access_type);
if (cmp == page) {
/* Found entry in victim tlb, swap tlb and iotlb. */
@@ -1507,7 +1507,7 @@ static int probe_access_internal(CPUArchState *env, target_ulong addr,
{
uintptr_t index = tlb_index(env, mmu_idx, addr);
CPUTLBEntry *entry = tlb_entry(env, mmu_idx, addr);
- target_ulong tlb_addr = tlb_read_idx(entry, access_type);
+ uint64_t tlb_addr = tlb_read_idx(entry, access_type);
target_ulong page_addr = addr & TARGET_PAGE_MASK;
int flags = TLB_FLAGS_MASK;
@@ -1694,7 +1694,7 @@ bool tlb_plugin_lookup(CPUState *cpu, vaddr addr, int mmu_idx,
CPUArchState *env = cpu->env_ptr;
CPUTLBEntry *tlbe = tlb_entry(env, mmu_idx, addr);
uintptr_t index = tlb_index(env, mmu_idx, addr);
- vaddr tlb_addr = is_store ? tlb_addr_write(tlbe) : tlbe->addr_read;
+ uint64_t tlb_addr = is_store ? tlb_addr_write(tlbe) : tlbe->addr_read;
if (likely(tlb_hit(tlb_addr, addr))) {
/* We must have an iotlb entry for MMIO */
@@ -1759,7 +1759,7 @@ static bool mmu_lookup1(CPUArchState *env, MMULookupPageData *data,
target_ulong addr = data->addr;
uintptr_t index = tlb_index(env, mmu_idx, addr);
CPUTLBEntry *entry = tlb_entry(env, mmu_idx, addr);
- target_ulong tlb_addr = tlb_read_idx(entry, access_type);
+ uint64_t tlb_addr = tlb_read_idx(entry, access_type);
bool maybe_resized = false;
/* If the TLB entry is for a different page, reload and try again. */
--
2.34.1
^ permalink raw reply related [flat|nested] 28+ messages in thread* [PULL 05/22] accel/tcg/cputlb.c: Widen addr in MMULookupPageData
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
` (3 preceding siblings ...)
2023-06-26 15:39 ` [PULL 04/22] accel/tcg/cputlb.c: Widen CPUTLBEntry access functions Richard Henderson
@ 2023-06-26 15:39 ` Richard Henderson
2023-06-26 15:39 ` [PULL 06/22] accel/tcg/cpu-exec.c: Widen pc to vaddr Richard Henderson
` (17 subsequent siblings)
22 siblings, 0 replies; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 15:39 UTC (permalink / raw)
To: qemu-devel; +Cc: Anton Johansson
From: Anton Johansson <anjo@rev.ng>
Functions accessing MMULookupPageData are also updated.
Signed-off-by: Anton Johansson <anjo@rev.ng>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230621135633.1649-6-anjo@rev.ng>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
accel/tcg/cputlb.c | 30 +++++++++++++++---------------
1 file changed, 15 insertions(+), 15 deletions(-)
diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c
index ac990a1526..cc53d0fb64 100644
--- a/accel/tcg/cputlb.c
+++ b/accel/tcg/cputlb.c
@@ -1729,7 +1729,7 @@ bool tlb_plugin_lookup(CPUState *cpu, vaddr addr, int mmu_idx,
typedef struct MMULookupPageData {
CPUTLBEntryFull *full;
void *haddr;
- target_ulong addr;
+ vaddr addr;
int flags;
int size;
} MMULookupPageData;
@@ -1756,7 +1756,7 @@ typedef struct MMULookupLocals {
static bool mmu_lookup1(CPUArchState *env, MMULookupPageData *data,
int mmu_idx, MMUAccessType access_type, uintptr_t ra)
{
- target_ulong addr = data->addr;
+ vaddr addr = data->addr;
uintptr_t index = tlb_index(env, mmu_idx, addr);
CPUTLBEntry *entry = tlb_entry(env, mmu_idx, addr);
uint64_t tlb_addr = tlb_read_idx(entry, access_type);
@@ -1796,7 +1796,7 @@ static void mmu_watch_or_dirty(CPUArchState *env, MMULookupPageData *data,
MMUAccessType access_type, uintptr_t ra)
{
CPUTLBEntryFull *full = data->full;
- target_ulong addr = data->addr;
+ vaddr addr = data->addr;
int flags = data->flags;
int size = data->size;
@@ -1827,7 +1827,7 @@ static void mmu_watch_or_dirty(CPUArchState *env, MMULookupPageData *data,
* Resolve the translation for the page(s) beginning at @addr, for MemOp.size
* bytes. Return true if the lookup crosses a page boundary.
*/
-static bool mmu_lookup(CPUArchState *env, target_ulong addr, MemOpIdx oi,
+static bool mmu_lookup(CPUArchState *env, vaddr addr, MemOpIdx oi,
uintptr_t ra, MMUAccessType type, MMULookupLocals *l)
{
unsigned a_bits;
@@ -2024,7 +2024,7 @@ static uint64_t do_ld_mmio_beN(CPUArchState *env, MMULookupPageData *p,
MMUAccessType type, uintptr_t ra)
{
CPUTLBEntryFull *full = p->full;
- target_ulong addr = p->addr;
+ vaddr addr = p->addr;
int i, size = p->size;
QEMU_IOTHREAD_LOCK_GUARD();
@@ -2333,7 +2333,7 @@ static uint64_t do_ld_8(CPUArchState *env, MMULookupPageData *p, int mmu_idx,
return ret;
}
-static uint8_t do_ld1_mmu(CPUArchState *env, target_ulong addr, MemOpIdx oi,
+static uint8_t do_ld1_mmu(CPUArchState *env, vaddr addr, MemOpIdx oi,
uintptr_t ra, MMUAccessType access_type)
{
MMULookupLocals l;
@@ -2352,7 +2352,7 @@ tcg_target_ulong helper_ldub_mmu(CPUArchState *env, uint64_t addr,
return do_ld1_mmu(env, addr, oi, retaddr, MMU_DATA_LOAD);
}
-static uint16_t do_ld2_mmu(CPUArchState *env, target_ulong addr, MemOpIdx oi,
+static uint16_t do_ld2_mmu(CPUArchState *env, vaddr addr, MemOpIdx oi,
uintptr_t ra, MMUAccessType access_type)
{
MMULookupLocals l;
@@ -2383,7 +2383,7 @@ tcg_target_ulong helper_lduw_mmu(CPUArchState *env, uint64_t addr,
return do_ld2_mmu(env, addr, oi, retaddr, MMU_DATA_LOAD);
}
-static uint32_t do_ld4_mmu(CPUArchState *env, target_ulong addr, MemOpIdx oi,
+static uint32_t do_ld4_mmu(CPUArchState *env, vaddr addr, MemOpIdx oi,
uintptr_t ra, MMUAccessType access_type)
{
MMULookupLocals l;
@@ -2410,7 +2410,7 @@ tcg_target_ulong helper_ldul_mmu(CPUArchState *env, uint64_t addr,
return do_ld4_mmu(env, addr, oi, retaddr, MMU_DATA_LOAD);
}
-static uint64_t do_ld8_mmu(CPUArchState *env, target_ulong addr, MemOpIdx oi,
+static uint64_t do_ld8_mmu(CPUArchState *env, vaddr addr, MemOpIdx oi,
uintptr_t ra, MMUAccessType access_type)
{
MMULookupLocals l;
@@ -2460,7 +2460,7 @@ tcg_target_ulong helper_ldsl_mmu(CPUArchState *env, uint64_t addr,
return (int32_t)helper_ldul_mmu(env, addr, oi, retaddr);
}
-static Int128 do_ld16_mmu(CPUArchState *env, target_ulong addr,
+static Int128 do_ld16_mmu(CPUArchState *env, vaddr addr,
MemOpIdx oi, uintptr_t ra)
{
MMULookupLocals l;
@@ -2617,7 +2617,7 @@ static uint64_t do_st_mmio_leN(CPUArchState *env, MMULookupPageData *p,
uint64_t val_le, int mmu_idx, uintptr_t ra)
{
CPUTLBEntryFull *full = p->full;
- target_ulong addr = p->addr;
+ vaddr addr = p->addr;
int i, size = p->size;
QEMU_IOTHREAD_LOCK_GUARD();
@@ -2808,7 +2808,7 @@ void helper_stb_mmu(CPUArchState *env, uint64_t addr, uint32_t val,
do_st_1(env, &l.page[0], val, l.mmu_idx, ra);
}
-static void do_st2_mmu(CPUArchState *env, target_ulong addr, uint16_t val,
+static void do_st2_mmu(CPUArchState *env, vaddr addr, uint16_t val,
MemOpIdx oi, uintptr_t ra)
{
MMULookupLocals l;
@@ -2837,7 +2837,7 @@ void helper_stw_mmu(CPUArchState *env, uint64_t addr, uint32_t val,
do_st2_mmu(env, addr, val, oi, retaddr);
}
-static void do_st4_mmu(CPUArchState *env, target_ulong addr, uint32_t val,
+static void do_st4_mmu(CPUArchState *env, vaddr addr, uint32_t val,
MemOpIdx oi, uintptr_t ra)
{
MMULookupLocals l;
@@ -2864,7 +2864,7 @@ void helper_stl_mmu(CPUArchState *env, uint64_t addr, uint32_t val,
do_st4_mmu(env, addr, val, oi, retaddr);
}
-static void do_st8_mmu(CPUArchState *env, target_ulong addr, uint64_t val,
+static void do_st8_mmu(CPUArchState *env, vaddr addr, uint64_t val,
MemOpIdx oi, uintptr_t ra)
{
MMULookupLocals l;
@@ -2891,7 +2891,7 @@ void helper_stq_mmu(CPUArchState *env, uint64_t addr, uint64_t val,
do_st8_mmu(env, addr, val, oi, retaddr);
}
-static void do_st16_mmu(CPUArchState *env, target_ulong addr, Int128 val,
+static void do_st16_mmu(CPUArchState *env, vaddr addr, Int128 val,
MemOpIdx oi, uintptr_t ra)
{
MMULookupLocals l;
--
2.34.1
^ permalink raw reply related [flat|nested] 28+ messages in thread* [PULL 06/22] accel/tcg/cpu-exec.c: Widen pc to vaddr
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
` (4 preceding siblings ...)
2023-06-26 15:39 ` [PULL 05/22] accel/tcg/cputlb.c: Widen addr in MMULookupPageData Richard Henderson
@ 2023-06-26 15:39 ` Richard Henderson
2023-07-11 16:40 ` Peter Maydell
2023-06-26 15:39 ` [PULL 07/22] accel/tcg: Widen pc to vaddr in CPUJumpCache Richard Henderson
` (16 subsequent siblings)
22 siblings, 1 reply; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 15:39 UTC (permalink / raw)
To: qemu-devel; +Cc: Anton Johansson
From: Anton Johansson <anjo@rev.ng>
Signed-off-by: Anton Johansson <anjo@rev.ng>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230621135633.1649-7-anjo@rev.ng>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
accel/tcg/cpu-exec.c | 34 +++++++++++++++++-----------------
1 file changed, 17 insertions(+), 17 deletions(-)
diff --git a/accel/tcg/cpu-exec.c b/accel/tcg/cpu-exec.c
index 4d952a6cc2..ba1890a373 100644
--- a/accel/tcg/cpu-exec.c
+++ b/accel/tcg/cpu-exec.c
@@ -169,8 +169,8 @@ uint32_t curr_cflags(CPUState *cpu)
}
struct tb_desc {
- target_ulong pc;
- target_ulong cs_base;
+ vaddr pc;
+ uint64_t cs_base;
CPUArchState *env;
tb_page_addr_t page_addr0;
uint32_t flags;
@@ -193,7 +193,7 @@ static bool tb_lookup_cmp(const void *p, const void *d)
return true;
} else {
tb_page_addr_t phys_page1;
- target_ulong virt_page1;
+ vaddr virt_page1;
/*
* We know that the first page matched, and an otherwise valid TB
@@ -214,8 +214,8 @@ static bool tb_lookup_cmp(const void *p, const void *d)
return false;
}
-static TranslationBlock *tb_htable_lookup(CPUState *cpu, target_ulong pc,
- target_ulong cs_base, uint32_t flags,
+static TranslationBlock *tb_htable_lookup(CPUState *cpu, vaddr pc,
+ uint64_t cs_base, uint32_t flags,
uint32_t cflags)
{
tb_page_addr_t phys_pc;
@@ -238,9 +238,9 @@ static TranslationBlock *tb_htable_lookup(CPUState *cpu, target_ulong pc,
}
/* Might cause an exception, so have a longjmp destination ready */
-static inline TranslationBlock *tb_lookup(CPUState *cpu, target_ulong pc,
- target_ulong cs_base,
- uint32_t flags, uint32_t cflags)
+static inline TranslationBlock *tb_lookup(CPUState *cpu, vaddr pc,
+ uint64_t cs_base, uint32_t flags,
+ uint32_t cflags)
{
TranslationBlock *tb;
CPUJumpCache *jc;
@@ -292,13 +292,13 @@ static inline TranslationBlock *tb_lookup(CPUState *cpu, target_ulong pc,
return tb;
}
-static void log_cpu_exec(target_ulong pc, CPUState *cpu,
+static void log_cpu_exec(vaddr pc, CPUState *cpu,
const TranslationBlock *tb)
{
if (qemu_log_in_addr_range(pc)) {
qemu_log_mask(CPU_LOG_EXEC,
"Trace %d: %p [%08" PRIx64
- "/" TARGET_FMT_lx "/%08x/%08x] %s\n",
+ "/%" VADDR_PRIx "/%08x/%08x] %s\n",
cpu->cpu_index, tb->tc.ptr, tb->cs_base, pc,
tb->flags, tb->cflags, lookup_symbol(pc));
@@ -323,7 +323,7 @@ static void log_cpu_exec(target_ulong pc, CPUState *cpu,
}
}
-static bool check_for_breakpoints_slow(CPUState *cpu, target_ulong pc,
+static bool check_for_breakpoints_slow(CPUState *cpu, vaddr pc,
uint32_t *cflags)
{
CPUBreakpoint *bp;
@@ -389,7 +389,7 @@ static bool check_for_breakpoints_slow(CPUState *cpu, target_ulong pc,
return false;
}
-static inline bool check_for_breakpoints(CPUState *cpu, target_ulong pc,
+static inline bool check_for_breakpoints(CPUState *cpu, vaddr pc,
uint32_t *cflags)
{
return unlikely(!QTAILQ_EMPTY(&cpu->breakpoints)) &&
@@ -485,10 +485,10 @@ cpu_tb_exec(CPUState *cpu, TranslationBlock *itb, int *tb_exit)
cc->set_pc(cpu, last_tb->pc);
}
if (qemu_loglevel_mask(CPU_LOG_EXEC)) {
- target_ulong pc = log_pc(cpu, last_tb);
+ vaddr pc = log_pc(cpu, last_tb);
if (qemu_log_in_addr_range(pc)) {
- qemu_log("Stopped execution of TB chain before %p ["
- TARGET_FMT_lx "] %s\n",
+ qemu_log("Stopped execution of TB chain before %p [%"
+ VADDR_PRIx "] %s\n",
last_tb->tc.ptr, pc, lookup_symbol(pc));
}
}
@@ -882,8 +882,8 @@ static inline bool cpu_handle_interrupt(CPUState *cpu,
}
static inline void cpu_loop_exec_tb(CPUState *cpu, TranslationBlock *tb,
- target_ulong pc,
- TranslationBlock **last_tb, int *tb_exit)
+ vaddr pc, TranslationBlock **last_tb,
+ int *tb_exit)
{
int32_t insns_left;
--
2.34.1
^ permalink raw reply related [flat|nested] 28+ messages in thread* Re: [PULL 06/22] accel/tcg/cpu-exec.c: Widen pc to vaddr
2023-06-26 15:39 ` [PULL 06/22] accel/tcg/cpu-exec.c: Widen pc to vaddr Richard Henderson
@ 2023-07-11 16:40 ` Peter Maydell
0 siblings, 0 replies; 28+ messages in thread
From: Peter Maydell @ 2023-07-11 16:40 UTC (permalink / raw)
To: Richard Henderson; +Cc: qemu-devel, Anton Johansson
On Mon, 26 Jun 2023 at 16:48, Richard Henderson
<richard.henderson@linaro.org> wrote:
>
> From: Anton Johansson <anjo@rev.ng>
>
> Signed-off-by: Anton Johansson <anjo@rev.ng>
> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
> Message-Id: <20230621135633.1649-7-anjo@rev.ng>
> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
> -static void log_cpu_exec(target_ulong pc, CPUState *cpu,
> +static void log_cpu_exec(vaddr pc, CPUState *cpu,
> const TranslationBlock *tb)
> {
> if (qemu_log_in_addr_range(pc)) {
> qemu_log_mask(CPU_LOG_EXEC,
> "Trace %d: %p [%08" PRIx64
> - "/" TARGET_FMT_lx "/%08x/%08x] %s\n",
> + "/%" VADDR_PRIx "/%08x/%08x] %s\n",
> cpu->cpu_index, tb->tc.ptr, tb->cs_base, pc,
> tb->flags, tb->cflags, lookup_symbol(pc));
This again has lost the zero-padding. I noticed this one because
I have a workflow where I post-process these log files to
extract the executed PC values.
We've also lost the "PC is padded to the appropriate size
depending on whether this is a 32-bit or 64-bit guest CPU",
which is a bit of a shame.
thanks
-- PMM
^ permalink raw reply [flat|nested] 28+ messages in thread
* [PULL 07/22] accel/tcg: Widen pc to vaddr in CPUJumpCache
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
` (5 preceding siblings ...)
2023-06-26 15:39 ` [PULL 06/22] accel/tcg/cpu-exec.c: Widen pc to vaddr Richard Henderson
@ 2023-06-26 15:39 ` Richard Henderson
2023-06-26 15:39 ` [PULL 08/22] accel: Replace target_ulong with vaddr in probe_*() Richard Henderson
` (15 subsequent siblings)
22 siblings, 0 replies; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 15:39 UTC (permalink / raw)
To: qemu-devel; +Cc: Anton Johansson
From: Anton Johansson <anjo@rev.ng>
Related functions dealing with the jump cache are also updated.
Signed-off-by: Anton Johansson <anjo@rev.ng>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230621135633.1649-8-anjo@rev.ng>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
accel/tcg/tb-hash.h | 12 ++++++------
accel/tcg/tb-jmp-cache.h | 2 +-
accel/tcg/cputlb.c | 2 +-
3 files changed, 8 insertions(+), 8 deletions(-)
diff --git a/accel/tcg/tb-hash.h b/accel/tcg/tb-hash.h
index 2ba2193731..a0c61f25cd 100644
--- a/accel/tcg/tb-hash.h
+++ b/accel/tcg/tb-hash.h
@@ -35,16 +35,16 @@
#define TB_JMP_ADDR_MASK (TB_JMP_PAGE_SIZE - 1)
#define TB_JMP_PAGE_MASK (TB_JMP_CACHE_SIZE - TB_JMP_PAGE_SIZE)
-static inline unsigned int tb_jmp_cache_hash_page(target_ulong pc)
+static inline unsigned int tb_jmp_cache_hash_page(vaddr pc)
{
- target_ulong tmp;
+ vaddr tmp;
tmp = pc ^ (pc >> (TARGET_PAGE_BITS - TB_JMP_PAGE_BITS));
return (tmp >> (TARGET_PAGE_BITS - TB_JMP_PAGE_BITS)) & TB_JMP_PAGE_MASK;
}
-static inline unsigned int tb_jmp_cache_hash_func(target_ulong pc)
+static inline unsigned int tb_jmp_cache_hash_func(vaddr pc)
{
- target_ulong tmp;
+ vaddr tmp;
tmp = pc ^ (pc >> (TARGET_PAGE_BITS - TB_JMP_PAGE_BITS));
return (((tmp >> (TARGET_PAGE_BITS - TB_JMP_PAGE_BITS)) & TB_JMP_PAGE_MASK)
| (tmp & TB_JMP_ADDR_MASK));
@@ -53,7 +53,7 @@ static inline unsigned int tb_jmp_cache_hash_func(target_ulong pc)
#else
/* In user-mode we can get better hashing because we do not have a TLB */
-static inline unsigned int tb_jmp_cache_hash_func(target_ulong pc)
+static inline unsigned int tb_jmp_cache_hash_func(vaddr pc)
{
return (pc ^ (pc >> TB_JMP_CACHE_BITS)) & (TB_JMP_CACHE_SIZE - 1);
}
@@ -61,7 +61,7 @@ static inline unsigned int tb_jmp_cache_hash_func(target_ulong pc)
#endif /* CONFIG_SOFTMMU */
static inline
-uint32_t tb_hash_func(tb_page_addr_t phys_pc, target_ulong pc,
+uint32_t tb_hash_func(tb_page_addr_t phys_pc, vaddr pc,
uint32_t flags, uint64_t flags2, uint32_t cf_mask)
{
return qemu_xxhash8(phys_pc, pc, flags2, flags, cf_mask);
diff --git a/accel/tcg/tb-jmp-cache.h b/accel/tcg/tb-jmp-cache.h
index bee87eb840..bb424c8a05 100644
--- a/accel/tcg/tb-jmp-cache.h
+++ b/accel/tcg/tb-jmp-cache.h
@@ -21,7 +21,7 @@ struct CPUJumpCache {
struct rcu_head rcu;
struct {
TranslationBlock *tb;
- target_ulong pc;
+ vaddr pc;
} array[TB_JMP_CACHE_SIZE];
};
diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c
index cc53d0fb64..bdf400f6e6 100644
--- a/accel/tcg/cputlb.c
+++ b/accel/tcg/cputlb.c
@@ -99,7 +99,7 @@ static void tlb_window_reset(CPUTLBDesc *desc, int64_t ns,
desc->window_max_entries = max_entries;
}
-static void tb_jmp_cache_clear_page(CPUState *cpu, target_ulong page_addr)
+static void tb_jmp_cache_clear_page(CPUState *cpu, vaddr page_addr)
{
CPUJumpCache *jc = cpu->tb_jmp_cache;
int i, i0;
--
2.34.1
^ permalink raw reply related [flat|nested] 28+ messages in thread* [PULL 08/22] accel: Replace target_ulong with vaddr in probe_*()
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
` (6 preceding siblings ...)
2023-06-26 15:39 ` [PULL 07/22] accel/tcg: Widen pc to vaddr in CPUJumpCache Richard Henderson
@ 2023-06-26 15:39 ` Richard Henderson
2023-06-26 15:39 ` [PULL 09/22] accel/tcg: Replace target_ulong with vaddr in *_mmu_lookup() Richard Henderson
` (14 subsequent siblings)
22 siblings, 0 replies; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 15:39 UTC (permalink / raw)
To: qemu-devel; +Cc: Anton Johansson
From: Anton Johansson <anjo@rev.ng>
Functions for probing memory accesses (and functions that call these)
are updated to take a vaddr for guest virtual addresses over
target_ulong.
Signed-off-by: Anton Johansson <anjo@rev.ng>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230621135633.1649-9-anjo@rev.ng>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
include/exec/exec-all.h | 14 +++++++-------
accel/stubs/tcg-stub.c | 4 ++--
accel/tcg/cputlb.c | 12 ++++++------
accel/tcg/user-exec.c | 8 ++++----
4 files changed, 19 insertions(+), 19 deletions(-)
diff --git a/include/exec/exec-all.h b/include/exec/exec-all.h
index f5508e242b..cc1c3556f6 100644
--- a/include/exec/exec-all.h
+++ b/include/exec/exec-all.h
@@ -413,16 +413,16 @@ static inline void tlb_flush_range_by_mmuidx_all_cpus_synced(CPUState *cpu,
* Finally, return the host address for a page that is backed by RAM,
* or NULL if the page requires I/O.
*/
-void *probe_access(CPUArchState *env, target_ulong addr, int size,
+void *probe_access(CPUArchState *env, vaddr addr, int size,
MMUAccessType access_type, int mmu_idx, uintptr_t retaddr);
-static inline void *probe_write(CPUArchState *env, target_ulong addr, int size,
+static inline void *probe_write(CPUArchState *env, vaddr addr, int size,
int mmu_idx, uintptr_t retaddr)
{
return probe_access(env, addr, size, MMU_DATA_STORE, mmu_idx, retaddr);
}
-static inline void *probe_read(CPUArchState *env, target_ulong addr, int size,
+static inline void *probe_read(CPUArchState *env, vaddr addr, int size,
int mmu_idx, uintptr_t retaddr)
{
return probe_access(env, addr, size, MMU_DATA_LOAD, mmu_idx, retaddr);
@@ -447,7 +447,7 @@ static inline void *probe_read(CPUArchState *env, target_ulong addr, int size,
* Do handle clean pages, so exclude TLB_NOTDIRY from the returned flags.
* For simplicity, all "mmio-like" flags are folded to TLB_MMIO.
*/
-int probe_access_flags(CPUArchState *env, target_ulong addr, int size,
+int probe_access_flags(CPUArchState *env, vaddr addr, int size,
MMUAccessType access_type, int mmu_idx,
bool nonfault, void **phost, uintptr_t retaddr);
@@ -460,7 +460,7 @@ int probe_access_flags(CPUArchState *env, target_ulong addr, int size,
* and must be consumed or copied immediately, before any further
* access or changes to TLB @mmu_idx.
*/
-int probe_access_full(CPUArchState *env, target_ulong addr, int size,
+int probe_access_full(CPUArchState *env, vaddr addr, int size,
MMUAccessType access_type, int mmu_idx,
bool nonfault, void **phost,
CPUTLBEntryFull **pfull, uintptr_t retaddr);
@@ -581,7 +581,7 @@ struct MemoryRegionSection *iotlb_to_section(CPUState *cpu,
*
* Note: this function can trigger an exception.
*/
-tb_page_addr_t get_page_addr_code_hostp(CPUArchState *env, target_ulong addr,
+tb_page_addr_t get_page_addr_code_hostp(CPUArchState *env, vaddr addr,
void **hostp);
/**
@@ -596,7 +596,7 @@ tb_page_addr_t get_page_addr_code_hostp(CPUArchState *env, target_ulong addr,
* Note: this function can trigger an exception.
*/
static inline tb_page_addr_t get_page_addr_code(CPUArchState *env,
- target_ulong addr)
+ vaddr addr)
{
return get_page_addr_code_hostp(env, addr, NULL);
}
diff --git a/accel/stubs/tcg-stub.c b/accel/stubs/tcg-stub.c
index 0998e601ad..a9e7a2d5b4 100644
--- a/accel/stubs/tcg-stub.c
+++ b/accel/stubs/tcg-stub.c
@@ -26,14 +26,14 @@ void tcg_flush_jmp_cache(CPUState *cpu)
{
}
-int probe_access_flags(CPUArchState *env, target_ulong addr, int size,
+int probe_access_flags(CPUArchState *env, vaddr addr, int size,
MMUAccessType access_type, int mmu_idx,
bool nonfault, void **phost, uintptr_t retaddr)
{
g_assert_not_reached();
}
-void *probe_access(CPUArchState *env, target_ulong addr, int size,
+void *probe_access(CPUArchState *env, vaddr addr, int size,
MMUAccessType access_type, int mmu_idx, uintptr_t retaddr)
{
/* Handled by hardware accelerator. */
diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c
index bdf400f6e6..d873e58a5d 100644
--- a/accel/tcg/cputlb.c
+++ b/accel/tcg/cputlb.c
@@ -1499,7 +1499,7 @@ static void notdirty_write(CPUState *cpu, vaddr mem_vaddr, unsigned size,
}
}
-static int probe_access_internal(CPUArchState *env, target_ulong addr,
+static int probe_access_internal(CPUArchState *env, vaddr addr,
int fault_size, MMUAccessType access_type,
int mmu_idx, bool nonfault,
void **phost, CPUTLBEntryFull **pfull,
@@ -1508,7 +1508,7 @@ static int probe_access_internal(CPUArchState *env, target_ulong addr,
uintptr_t index = tlb_index(env, mmu_idx, addr);
CPUTLBEntry *entry = tlb_entry(env, mmu_idx, addr);
uint64_t tlb_addr = tlb_read_idx(entry, access_type);
- target_ulong page_addr = addr & TARGET_PAGE_MASK;
+ vaddr page_addr = addr & TARGET_PAGE_MASK;
int flags = TLB_FLAGS_MASK;
if (!tlb_hit_page(tlb_addr, page_addr)) {
@@ -1551,7 +1551,7 @@ static int probe_access_internal(CPUArchState *env, target_ulong addr,
return flags;
}
-int probe_access_full(CPUArchState *env, target_ulong addr, int size,
+int probe_access_full(CPUArchState *env, vaddr addr, int size,
MMUAccessType access_type, int mmu_idx,
bool nonfault, void **phost, CPUTLBEntryFull **pfull,
uintptr_t retaddr)
@@ -1568,7 +1568,7 @@ int probe_access_full(CPUArchState *env, target_ulong addr, int size,
return flags;
}
-int probe_access_flags(CPUArchState *env, target_ulong addr, int size,
+int probe_access_flags(CPUArchState *env, vaddr addr, int size,
MMUAccessType access_type, int mmu_idx,
bool nonfault, void **phost, uintptr_t retaddr)
{
@@ -1589,7 +1589,7 @@ int probe_access_flags(CPUArchState *env, target_ulong addr, int size,
return flags;
}
-void *probe_access(CPUArchState *env, target_ulong addr, int size,
+void *probe_access(CPUArchState *env, vaddr addr, int size,
MMUAccessType access_type, int mmu_idx, uintptr_t retaddr)
{
CPUTLBEntryFull *full;
@@ -1648,7 +1648,7 @@ void *tlb_vaddr_to_host(CPUArchState *env, abi_ptr addr,
* NOTE: This function will trigger an exception if the page is
* not executable.
*/
-tb_page_addr_t get_page_addr_code_hostp(CPUArchState *env, target_ulong addr,
+tb_page_addr_t get_page_addr_code_hostp(CPUArchState *env, vaddr addr,
void **hostp)
{
CPUTLBEntryFull *full;
diff --git a/accel/tcg/user-exec.c b/accel/tcg/user-exec.c
index dc8d6b5d40..d71e26a7b5 100644
--- a/accel/tcg/user-exec.c
+++ b/accel/tcg/user-exec.c
@@ -721,7 +721,7 @@ int page_unprotect(target_ulong address, uintptr_t pc)
return current_tb_invalidated ? 2 : 1;
}
-static int probe_access_internal(CPUArchState *env, target_ulong addr,
+static int probe_access_internal(CPUArchState *env, vaddr addr,
int fault_size, MMUAccessType access_type,
bool nonfault, uintptr_t ra)
{
@@ -759,7 +759,7 @@ static int probe_access_internal(CPUArchState *env, target_ulong addr,
cpu_loop_exit_sigsegv(env_cpu(env), addr, access_type, maperr, ra);
}
-int probe_access_flags(CPUArchState *env, target_ulong addr, int size,
+int probe_access_flags(CPUArchState *env, vaddr addr, int size,
MMUAccessType access_type, int mmu_idx,
bool nonfault, void **phost, uintptr_t ra)
{
@@ -771,7 +771,7 @@ int probe_access_flags(CPUArchState *env, target_ulong addr, int size,
return flags;
}
-void *probe_access(CPUArchState *env, target_ulong addr, int size,
+void *probe_access(CPUArchState *env, vaddr addr, int size,
MMUAccessType access_type, int mmu_idx, uintptr_t ra)
{
int flags;
@@ -783,7 +783,7 @@ void *probe_access(CPUArchState *env, target_ulong addr, int size,
return size ? g2h(env_cpu(env), addr) : NULL;
}
-tb_page_addr_t get_page_addr_code_hostp(CPUArchState *env, target_ulong addr,
+tb_page_addr_t get_page_addr_code_hostp(CPUArchState *env, vaddr addr,
void **hostp)
{
int flags;
--
2.34.1
^ permalink raw reply related [flat|nested] 28+ messages in thread* [PULL 09/22] accel/tcg: Replace target_ulong with vaddr in *_mmu_lookup()
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
` (7 preceding siblings ...)
2023-06-26 15:39 ` [PULL 08/22] accel: Replace target_ulong with vaddr in probe_*() Richard Henderson
@ 2023-06-26 15:39 ` Richard Henderson
2023-06-26 15:39 ` [PULL 10/22] accel/tcg: Replace target_ulong with vaddr in translator_*() Richard Henderson
` (13 subsequent siblings)
22 siblings, 0 replies; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 15:39 UTC (permalink / raw)
To: qemu-devel; +Cc: Anton Johansson
From: Anton Johansson <anjo@rev.ng>
Update atomic_mmu_lookup() and cpu_mmu_lookup() to take the guest
virtual address as a vaddr instead of a target_ulong.
Signed-off-by: Anton Johansson <anjo@rev.ng>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230621135633.1649-10-anjo@rev.ng>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
accel/tcg/cputlb.c | 6 +++---
accel/tcg/user-exec.c | 6 +++---
2 files changed, 6 insertions(+), 6 deletions(-)
diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c
index d873e58a5d..e02cfc550e 100644
--- a/accel/tcg/cputlb.c
+++ b/accel/tcg/cputlb.c
@@ -1898,15 +1898,15 @@ static bool mmu_lookup(CPUArchState *env, vaddr addr, MemOpIdx oi,
* Probe for an atomic operation. Do not allow unaligned operations,
* or io operations to proceed. Return the host address.
*/
-static void *atomic_mmu_lookup(CPUArchState *env, target_ulong addr,
- MemOpIdx oi, int size, uintptr_t retaddr)
+static void *atomic_mmu_lookup(CPUArchState *env, vaddr addr, MemOpIdx oi,
+ int size, uintptr_t retaddr)
{
uintptr_t mmu_idx = get_mmuidx(oi);
MemOp mop = get_memop(oi);
int a_bits = get_alignment_bits(mop);
uintptr_t index;
CPUTLBEntry *tlbe;
- target_ulong tlb_addr;
+ vaddr tlb_addr;
void *hostaddr;
CPUTLBEntryFull *full;
diff --git a/accel/tcg/user-exec.c b/accel/tcg/user-exec.c
index d71e26a7b5..f8b16d6ab8 100644
--- a/accel/tcg/user-exec.c
+++ b/accel/tcg/user-exec.c
@@ -889,7 +889,7 @@ void page_reset_target_data(target_ulong start, target_ulong last) { }
/* The softmmu versions of these helpers are in cputlb.c. */
-static void *cpu_mmu_lookup(CPUArchState *env, abi_ptr addr,
+static void *cpu_mmu_lookup(CPUArchState *env, vaddr addr,
MemOp mop, uintptr_t ra, MMUAccessType type)
{
int a_bits = get_alignment_bits(mop);
@@ -1324,8 +1324,8 @@ uint64_t cpu_ldq_code_mmu(CPUArchState *env, abi_ptr addr,
/*
* Do not allow unaligned operations to proceed. Return the host address.
*/
-static void *atomic_mmu_lookup(CPUArchState *env, target_ulong addr,
- MemOpIdx oi, int size, uintptr_t retaddr)
+static void *atomic_mmu_lookup(CPUArchState *env, vaddr addr, MemOpIdx oi,
+ int size, uintptr_t retaddr)
{
MemOp mop = get_memop(oi);
int a_bits = get_alignment_bits(mop);
--
2.34.1
^ permalink raw reply related [flat|nested] 28+ messages in thread* [PULL 10/22] accel/tcg: Replace target_ulong with vaddr in translator_*()
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
` (8 preceding siblings ...)
2023-06-26 15:39 ` [PULL 09/22] accel/tcg: Replace target_ulong with vaddr in *_mmu_lookup() Richard Henderson
@ 2023-06-26 15:39 ` Richard Henderson
2023-06-26 15:39 ` [PULL 11/22] cpu: Replace target_ulong with hwaddr in tb_invalidate_phys_addr() Richard Henderson
` (12 subsequent siblings)
22 siblings, 0 replies; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 15:39 UTC (permalink / raw)
To: qemu-devel; +Cc: Anton Johansson
From: Anton Johansson <anjo@rev.ng>
Use vaddr for guest virtual address in translator_use_goto_tb() and
translator_loop().
Signed-off-by: Anton Johansson <anjo@rev.ng>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230621135633.1649-11-anjo@rev.ng>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
include/exec/translator.h | 6 +++---
accel/tcg/translator.c | 10 +++++-----
2 files changed, 8 insertions(+), 8 deletions(-)
diff --git a/include/exec/translator.h b/include/exec/translator.h
index 224ae14aa7..a53d3243d4 100644
--- a/include/exec/translator.h
+++ b/include/exec/translator.h
@@ -142,8 +142,8 @@ typedef struct TranslatorOps {
* - When too many instructions have been translated.
*/
void translator_loop(CPUState *cpu, TranslationBlock *tb, int *max_insns,
- target_ulong pc, void *host_pc,
- const TranslatorOps *ops, DisasContextBase *db);
+ vaddr pc, void *host_pc, const TranslatorOps *ops,
+ DisasContextBase *db);
/**
* translator_use_goto_tb
@@ -153,7 +153,7 @@ void translator_loop(CPUState *cpu, TranslationBlock *tb, int *max_insns,
* Return true if goto_tb is allowed between the current TB
* and the destination PC.
*/
-bool translator_use_goto_tb(DisasContextBase *db, target_ulong dest);
+bool translator_use_goto_tb(DisasContextBase *db, vaddr dest);
/**
* translator_io_start
diff --git a/accel/tcg/translator.c b/accel/tcg/translator.c
index 918a455e73..0fd9efceba 100644
--- a/accel/tcg/translator.c
+++ b/accel/tcg/translator.c
@@ -117,7 +117,7 @@ static void gen_tb_end(const TranslationBlock *tb, uint32_t cflags,
}
}
-bool translator_use_goto_tb(DisasContextBase *db, target_ulong dest)
+bool translator_use_goto_tb(DisasContextBase *db, vaddr dest)
{
/* Suppress goto_tb if requested. */
if (tb_cflags(db->tb) & CF_NO_GOTO_TB) {
@@ -129,8 +129,8 @@ bool translator_use_goto_tb(DisasContextBase *db, target_ulong dest)
}
void translator_loop(CPUState *cpu, TranslationBlock *tb, int *max_insns,
- target_ulong pc, void *host_pc,
- const TranslatorOps *ops, DisasContextBase *db)
+ vaddr pc, void *host_pc, const TranslatorOps *ops,
+ DisasContextBase *db)
{
uint32_t cflags = tb_cflags(tb);
TCGOp *icount_start_insn;
@@ -235,10 +235,10 @@ void translator_loop(CPUState *cpu, TranslationBlock *tb, int *max_insns,
}
static void *translator_access(CPUArchState *env, DisasContextBase *db,
- target_ulong pc, size_t len)
+ vaddr pc, size_t len)
{
void *host;
- target_ulong base, end;
+ vaddr base, end;
TranslationBlock *tb;
tb = db->tb;
--
2.34.1
^ permalink raw reply related [flat|nested] 28+ messages in thread* [PULL 11/22] cpu: Replace target_ulong with hwaddr in tb_invalidate_phys_addr()
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
` (9 preceding siblings ...)
2023-06-26 15:39 ` [PULL 10/22] accel/tcg: Replace target_ulong with vaddr in translator_*() Richard Henderson
@ 2023-06-26 15:39 ` Richard Henderson
2023-06-26 15:39 ` [PULL 12/22] softfloat: use QEMU_FLATTEN to avoid mistaken isra inlining Richard Henderson
` (11 subsequent siblings)
22 siblings, 0 replies; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 15:39 UTC (permalink / raw)
To: qemu-devel; +Cc: Anton Johansson
From: Anton Johansson <anjo@rev.ng>
Signed-off-by: Anton Johansson <anjo@rev.ng>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230621135633.1649-13-anjo@rev.ng>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
include/exec/exec-all.h | 2 +-
cpu.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/include/exec/exec-all.h b/include/exec/exec-all.h
index cc1c3556f6..200c27eadf 100644
--- a/include/exec/exec-all.h
+++ b/include/exec/exec-all.h
@@ -526,7 +526,7 @@ uint32_t curr_cflags(CPUState *cpu);
/* TranslationBlock invalidate API */
#if defined(CONFIG_USER_ONLY)
-void tb_invalidate_phys_addr(target_ulong addr);
+void tb_invalidate_phys_addr(hwaddr addr);
#else
void tb_invalidate_phys_addr(AddressSpace *as, hwaddr addr, MemTxAttrs attrs);
#endif
diff --git a/cpu.c b/cpu.c
index 65ebaf8159..1c948d1161 100644
--- a/cpu.c
+++ b/cpu.c
@@ -293,7 +293,7 @@ void list_cpus(void)
}
#if defined(CONFIG_USER_ONLY)
-void tb_invalidate_phys_addr(target_ulong addr)
+void tb_invalidate_phys_addr(hwaddr addr)
{
mmap_lock();
tb_invalidate_phys_page(addr);
--
2.34.1
^ permalink raw reply related [flat|nested] 28+ messages in thread* [PULL 12/22] softfloat: use QEMU_FLATTEN to avoid mistaken isra inlining
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
` (10 preceding siblings ...)
2023-06-26 15:39 ` [PULL 11/22] cpu: Replace target_ulong with hwaddr in tb_invalidate_phys_addr() Richard Henderson
@ 2023-06-26 15:39 ` Richard Henderson
2023-06-26 15:39 ` [PULL 13/22] tests/plugin: Remove duplicate insn log from libinsn.so Richard Henderson
` (10 subsequent siblings)
22 siblings, 0 replies; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 15:39 UTC (permalink / raw)
To: qemu-devel; +Cc: Alex Bennée, BALATON Zoltan, Philippe Mathieu-Daudé
From: Alex Bennée <alex.bennee@linaro.org>
Balton discovered that asserts for the extract/deposit calls had a
significant impact on a lame benchmark on qemu-ppc. Replicating with:
./qemu-ppc64 ~/lsrc/tests/lame.git-svn/builds/ppc64/frontend/lame \
-h pts-trondheim-3.wav pts-trondheim-3.mp3
showed up the pack/unpack routines not eliding the assert checks as it
should have done causing them to prominently figure in the profile:
11.44% qemu-ppc64 qemu-ppc64 [.] unpack_raw64.isra.0
11.03% qemu-ppc64 qemu-ppc64 [.] parts64_uncanon_normal
8.26% qemu-ppc64 qemu-ppc64 [.] helper_compute_fprf_float64
6.75% qemu-ppc64 qemu-ppc64 [.] do_float_check_status
5.34% qemu-ppc64 qemu-ppc64 [.] parts64_muladd
4.75% qemu-ppc64 qemu-ppc64 [.] pack_raw64.isra.0
4.38% qemu-ppc64 qemu-ppc64 [.] parts64_canonicalize
3.62% qemu-ppc64 qemu-ppc64 [.] float64r32_round_pack_canonical
After this patch the same test runs 31 seconds faster with a profile
where the generated code dominates more:
+ 14.12% 0.00% qemu-ppc64 [unknown] [.] 0x0000004000619420
+ 13.30% 0.00% qemu-ppc64 [unknown] [.] 0x0000004000616850
+ 12.58% 12.19% qemu-ppc64 qemu-ppc64 [.] parts64_uncanon_normal
+ 10.62% 0.00% qemu-ppc64 [unknown] [.] 0x000000400061bf70
+ 9.91% 9.73% qemu-ppc64 qemu-ppc64 [.] helper_compute_fprf_float64
+ 7.84% 7.82% qemu-ppc64 qemu-ppc64 [.] do_float_check_status
+ 6.47% 5.78% qemu-ppc64 qemu-ppc64 [.] parts64_canonicalize.constprop.0
+ 6.46% 0.00% qemu-ppc64 [unknown] [.] 0x0000004000620130
+ 6.42% 0.00% qemu-ppc64 [unknown] [.] 0x0000004000619400
+ 6.17% 6.04% qemu-ppc64 qemu-ppc64 [.] parts64_muladd
+ 5.85% 0.00% qemu-ppc64 [unknown] [.] 0x00000040006167e0
+ 5.74% 0.00% qemu-ppc64 [unknown] [.] 0x0000b693fcffffd3
+ 5.45% 4.78% qemu-ppc64 qemu-ppc64 [.] float64r32_round_pack_canonical
Suggested-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <ec9cfe5a-d5f2-466d-34dc-c35817e7e010@linaro.org>
[AJB: Patchified rth's suggestion]
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Cc: BALATON Zoltan <balaton@eik.bme.hu>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: BALATON Zoltan <balaton@eik.bme.hu>
Message-Id: <20230523131107.3680641-1-alex.bennee@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
fpu/softfloat.c | 22 +++++++++++-----------
1 file changed, 11 insertions(+), 11 deletions(-)
diff --git a/fpu/softfloat.c b/fpu/softfloat.c
index 108f9cb224..42e6c188b4 100644
--- a/fpu/softfloat.c
+++ b/fpu/softfloat.c
@@ -593,27 +593,27 @@ static void unpack_raw64(FloatParts64 *r, const FloatFmt *fmt, uint64_t raw)
};
}
-static inline void float16_unpack_raw(FloatParts64 *p, float16 f)
+static void QEMU_FLATTEN float16_unpack_raw(FloatParts64 *p, float16 f)
{
unpack_raw64(p, &float16_params, f);
}
-static inline void bfloat16_unpack_raw(FloatParts64 *p, bfloat16 f)
+static void QEMU_FLATTEN bfloat16_unpack_raw(FloatParts64 *p, bfloat16 f)
{
unpack_raw64(p, &bfloat16_params, f);
}
-static inline void float32_unpack_raw(FloatParts64 *p, float32 f)
+static void QEMU_FLATTEN float32_unpack_raw(FloatParts64 *p, float32 f)
{
unpack_raw64(p, &float32_params, f);
}
-static inline void float64_unpack_raw(FloatParts64 *p, float64 f)
+static void QEMU_FLATTEN float64_unpack_raw(FloatParts64 *p, float64 f)
{
unpack_raw64(p, &float64_params, f);
}
-static void floatx80_unpack_raw(FloatParts128 *p, floatx80 f)
+static void QEMU_FLATTEN floatx80_unpack_raw(FloatParts128 *p, floatx80 f)
{
*p = (FloatParts128) {
.cls = float_class_unclassified,
@@ -623,7 +623,7 @@ static void floatx80_unpack_raw(FloatParts128 *p, floatx80 f)
};
}
-static void float128_unpack_raw(FloatParts128 *p, float128 f)
+static void QEMU_FLATTEN float128_unpack_raw(FloatParts128 *p, float128 f)
{
const int f_size = float128_params.frac_size - 64;
const int e_size = float128_params.exp_size;
@@ -650,27 +650,27 @@ static uint64_t pack_raw64(const FloatParts64 *p, const FloatFmt *fmt)
return ret;
}
-static inline float16 float16_pack_raw(const FloatParts64 *p)
+static float16 QEMU_FLATTEN float16_pack_raw(const FloatParts64 *p)
{
return make_float16(pack_raw64(p, &float16_params));
}
-static inline bfloat16 bfloat16_pack_raw(const FloatParts64 *p)
+static bfloat16 QEMU_FLATTEN bfloat16_pack_raw(const FloatParts64 *p)
{
return pack_raw64(p, &bfloat16_params);
}
-static inline float32 float32_pack_raw(const FloatParts64 *p)
+static float32 QEMU_FLATTEN float32_pack_raw(const FloatParts64 *p)
{
return make_float32(pack_raw64(p, &float32_params));
}
-static inline float64 float64_pack_raw(const FloatParts64 *p)
+static float64 QEMU_FLATTEN float64_pack_raw(const FloatParts64 *p)
{
return make_float64(pack_raw64(p, &float64_params));
}
-static float128 float128_pack_raw(const FloatParts128 *p)
+static float128 QEMU_FLATTEN float128_pack_raw(const FloatParts128 *p)
{
const int f_size = float128_params.frac_size - 64;
const int e_size = float128_params.exp_size;
--
2.34.1
^ permalink raw reply related [flat|nested] 28+ messages in thread* [PULL 13/22] tests/plugin: Remove duplicate insn log from libinsn.so
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
` (11 preceding siblings ...)
2023-06-26 15:39 ` [PULL 12/22] softfloat: use QEMU_FLATTEN to avoid mistaken isra inlining Richard Henderson
@ 2023-06-26 15:39 ` Richard Henderson
2023-06-26 15:39 ` [PULL 14/22] accel/tcg: remove CONFIG_PROFILER Richard Henderson
` (9 subsequent siblings)
22 siblings, 0 replies; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 15:39 UTC (permalink / raw)
To: qemu-devel
This is a perfectly natural occurrence for x86 "rep movb",
where the "rep" prefix forms a counted loop of the one insn.
During the tests/tcg/multiarch/memory test, this logging is
triggered over 350000 times. Within the context of cross-i386-tci
build, which is already slow by nature, the logging is sufficient
to push the test into timeout.
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
tests/plugin/insn.c | 9 +--------
tests/tcg/i386/Makefile.softmmu-target | 9 ---------
tests/tcg/i386/Makefile.target | 6 ------
tests/tcg/x86_64/Makefile.softmmu-target | 9 ---------
4 files changed, 1 insertion(+), 32 deletions(-)
diff --git a/tests/plugin/insn.c b/tests/plugin/insn.c
index cd5ea5d4ae..9bd6e44f73 100644
--- a/tests/plugin/insn.c
+++ b/tests/plugin/insn.c
@@ -19,7 +19,6 @@ QEMU_PLUGIN_EXPORT int qemu_plugin_version = QEMU_PLUGIN_VERSION;
#define MAX_CPUS 8 /* lets not go nuts */
typedef struct {
- uint64_t last_pc;
uint64_t insn_count;
} InstructionCount;
@@ -51,13 +50,7 @@ static void vcpu_insn_exec_before(unsigned int cpu_index, void *udata)
{
unsigned int i = cpu_index % MAX_CPUS;
InstructionCount *c = &counts[i];
- uint64_t this_pc = GPOINTER_TO_UINT(udata);
- if (this_pc == c->last_pc) {
- g_autofree gchar *out = g_strdup_printf("detected repeat execution @ 0x%"
- PRIx64 "\n", this_pc);
- qemu_plugin_outs(out);
- }
- c->last_pc = this_pc;
+
c->insn_count++;
}
diff --git a/tests/tcg/i386/Makefile.softmmu-target b/tests/tcg/i386/Makefile.softmmu-target
index ed922d59c8..5266f2335a 100644
--- a/tests/tcg/i386/Makefile.softmmu-target
+++ b/tests/tcg/i386/Makefile.softmmu-target
@@ -33,14 +33,5 @@ EXTRA_RUNS+=$(MULTIARCH_RUNS)
memory: CFLAGS+=-DCHECK_UNALIGNED=1
-# non-inline runs will trigger the duplicate instruction heuristics in libinsn.so
-run-plugin-%-with-libinsn.so:
- $(call run-test, $@, \
- $(QEMU) -monitor none -display none \
- -chardev file$(COMMA)path=$@.out$(COMMA)id=output \
- -plugin ../../plugin/libinsn.so$(COMMA)inline=on \
- -d plugin -D $*-with-libinsn.so.pout \
- $(QEMU_OPTS) $*)
-
# Running
QEMU_OPTS+=-device isa-debugcon,chardev=output -device isa-debug-exit,iobase=0xf4,iosize=0x4 -kernel
diff --git a/tests/tcg/i386/Makefile.target b/tests/tcg/i386/Makefile.target
index 821822ed0c..f2ee7a4db7 100644
--- a/tests/tcg/i386/Makefile.target
+++ b/tests/tcg/i386/Makefile.target
@@ -63,12 +63,6 @@ else
SKIP_I386_TESTS+=test-i386-fprem
endif
-# non-inline runs will trigger the duplicate instruction heuristics in libinsn.so
-run-plugin-%-with-libinsn.so:
- $(call run-test, $@, $(QEMU) $(QEMU_OPTS) \
- -plugin ../../plugin/libinsn.so$(COMMA)inline=on \
- -d plugin -D $*-with-libinsn.so.pout $*)
-
# Update TESTS
I386_TESTS:=$(filter-out $(SKIP_I386_TESTS), $(ALL_X86_TESTS))
TESTS=$(MULTIARCH_TESTS) $(I386_TESTS)
diff --git a/tests/tcg/x86_64/Makefile.softmmu-target b/tests/tcg/x86_64/Makefile.softmmu-target
index 7207fee94c..1bd763f2e6 100644
--- a/tests/tcg/x86_64/Makefile.softmmu-target
+++ b/tests/tcg/x86_64/Makefile.softmmu-target
@@ -33,14 +33,5 @@ EXTRA_RUNS+=$(MULTIARCH_RUNS)
memory: CFLAGS+=-DCHECK_UNALIGNED=1
-# non-inline runs will trigger the duplicate instruction heuristics in libinsn.so
-run-plugin-%-with-libinsn.so:
- $(call run-test, $@, \
- $(QEMU) -monitor none -display none \
- -chardev file$(COMMA)path=$@.out$(COMMA)id=output \
- -plugin ../../plugin/libinsn.so$(COMMA)inline=on \
- -d plugin -D $*-with-libinsn.so.pout \
- $(QEMU_OPTS) $*)
-
# Running
QEMU_OPTS+=-device isa-debugcon,chardev=output -device isa-debug-exit,iobase=0xf4,iosize=0x4 -kernel
--
2.34.1
^ permalink raw reply related [flat|nested] 28+ messages in thread* [PULL 14/22] accel/tcg: remove CONFIG_PROFILER
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
` (12 preceding siblings ...)
2023-06-26 15:39 ` [PULL 13/22] tests/plugin: Remove duplicate insn log from libinsn.so Richard Henderson
@ 2023-06-26 15:39 ` Richard Henderson
2023-06-26 15:39 ` [PULL 15/22] tcg: Fix temporary variable in tcg_gen_gvec_andcs Richard Henderson
` (8 subsequent siblings)
22 siblings, 0 replies; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 15:39 UTC (permalink / raw)
To: qemu-devel; +Cc: Fei Wu, Vanderson M . do Rosario, Alex Bennée
From: Fei Wu <fei2.wu@intel.com>
TBStats will be introduced to replace CONFIG_PROFILER totally, here
remove all CONFIG_PROFILER related stuffs first.
Signed-off-by: Vanderson M. do Rosario <vandersonmr2@gmail.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Fei Wu <fei2.wu@intel.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230607122411.3394702-2-fei2.wu@intel.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
meson.build | 2 -
qapi/machine.json | 18 ---
include/qemu/timer.h | 9 --
include/tcg/tcg.h | 26 -----
accel/tcg/monitor.c | 31 -----
accel/tcg/tcg-accel-ops.c | 10 --
accel/tcg/translate-all.c | 33 ------
softmmu/runstate.c | 9 --
tcg/tcg.c | 214 ----------------------------------
tests/qtest/qmp-cmd-test.c | 3 -
hmp-commands-info.hx | 15 ---
meson_options.txt | 2 -
scripts/meson-buildoptions.sh | 3 -
13 files changed, 375 deletions(-)
diff --git a/meson.build b/meson.build
index 3e3d38badb..b409788832 100644
--- a/meson.build
+++ b/meson.build
@@ -2121,7 +2121,6 @@ if numa.found()
dependencies: numa))
endif
config_host_data.set('CONFIG_OPENGL', opengl.found())
-config_host_data.set('CONFIG_PROFILER', get_option('profiler'))
config_host_data.set('CONFIG_RBD', rbd.found())
config_host_data.set('CONFIG_RDMA', rdma.found())
config_host_data.set('CONFIG_SAFESTACK', get_option('safe_stack'))
@@ -4087,7 +4086,6 @@ if 'objc' in all_languages
summary_info += {'QEMU_OBJCFLAGS': ' '.join(qemu_common_flags)}
endif
summary_info += {'QEMU_LDFLAGS': ' '.join(qemu_ldflags)}
-summary_info += {'profiler': get_option('profiler')}
summary_info += {'link-time optimization (LTO)': get_option('b_lto')}
summary_info += {'PIE': get_option('b_pie')}
summary_info += {'static build': get_option('prefer_static')}
diff --git a/qapi/machine.json b/qapi/machine.json
index 37660d8f2a..a08b6576ca 100644
--- a/qapi/machine.json
+++ b/qapi/machine.json
@@ -1575,24 +1575,6 @@
'if': 'CONFIG_TCG',
'features': [ 'unstable' ] }
-##
-# @x-query-profile:
-#
-# Query TCG profiling information
-#
-# Features:
-#
-# @unstable: This command is meant for debugging.
-#
-# Returns: profile information
-#
-# Since: 6.2
-##
-{ 'command': 'x-query-profile',
- 'returns': 'HumanReadableText',
- 'if': 'CONFIG_TCG',
- 'features': [ 'unstable' ] }
-
##
# @x-query-ramblock:
#
diff --git a/include/qemu/timer.h b/include/qemu/timer.h
index ee071e07d1..9a91cb1248 100644
--- a/include/qemu/timer.h
+++ b/include/qemu/timer.h
@@ -989,13 +989,4 @@ static inline int64_t cpu_get_host_ticks(void)
}
#endif
-#ifdef CONFIG_PROFILER
-static inline int64_t profile_getclock(void)
-{
- return get_clock();
-}
-
-extern int64_t dev_time;
-#endif
-
#endif
diff --git a/include/tcg/tcg.h b/include/tcg/tcg.h
index a498f31967..95541e9474 100644
--- a/include/tcg/tcg.h
+++ b/include/tcg/tcg.h
@@ -478,27 +478,6 @@ static inline TCGRegSet output_pref(const TCGOp *op, unsigned i)
return i < ARRAY_SIZE(op->output_pref) ? op->output_pref[i] : 0;
}
-typedef struct TCGProfile {
- int64_t cpu_exec_time;
- int64_t tb_count1;
- int64_t tb_count;
- int64_t op_count; /* total insn count */
- int op_count_max; /* max insn per TB */
- int temp_count_max;
- int64_t temp_count;
- int64_t del_op_count;
- int64_t code_in_len;
- int64_t code_out_len;
- int64_t search_out_len;
- int64_t interm_time;
- int64_t code_time;
- int64_t la_time;
- int64_t opt_time;
- int64_t restore_count;
- int64_t restore_time;
- int64_t table_op_count[NB_OPS];
-} TCGProfile;
-
struct TCGContext {
uint8_t *pool_cur, *pool_end;
TCGPool *pool_first, *pool_current, *pool_first_large;
@@ -528,10 +507,6 @@ struct TCGContext {
tcg_insn_unit *code_buf; /* pointer for start of tb */
tcg_insn_unit *code_ptr; /* pointer for running end of tb */
-#ifdef CONFIG_PROFILER
- TCGProfile prof;
-#endif
-
#ifdef CONFIG_DEBUG_TCG
int goto_tb_issue_mask;
const TCGOpcode *vecop_list;
@@ -871,7 +846,6 @@ static inline TCGv_ptr tcg_temp_new_ptr(void)
return temp_tcgv_ptr(t);
}
-int64_t tcg_cpu_exec_time(void);
void tcg_dump_info(GString *buf);
void tcg_dump_op_count(GString *buf);
diff --git a/accel/tcg/monitor.c b/accel/tcg/monitor.c
index f171bc6f5e..d48de23999 100644
--- a/accel/tcg/monitor.c
+++ b/accel/tcg/monitor.c
@@ -81,37 +81,6 @@ HumanReadableText *qmp_x_query_opcount(Error **errp)
return human_readable_text_from_str(buf);
}
-#ifdef CONFIG_PROFILER
-
-int64_t dev_time;
-
-HumanReadableText *qmp_x_query_profile(Error **errp)
-{
- g_autoptr(GString) buf = g_string_new("");
- static int64_t last_cpu_exec_time;
- int64_t cpu_exec_time;
- int64_t delta;
-
- cpu_exec_time = tcg_cpu_exec_time();
- delta = cpu_exec_time - last_cpu_exec_time;
-
- g_string_append_printf(buf, "async time %" PRId64 " (%0.3f)\n",
- dev_time, dev_time / (double)NANOSECONDS_PER_SECOND);
- g_string_append_printf(buf, "qemu time %" PRId64 " (%0.3f)\n",
- delta, delta / (double)NANOSECONDS_PER_SECOND);
- last_cpu_exec_time = cpu_exec_time;
- dev_time = 0;
-
- return human_readable_text_from_str(buf);
-}
-#else
-HumanReadableText *qmp_x_query_profile(Error **errp)
-{
- error_setg(errp, "Internal profiler not compiled");
- return NULL;
-}
-#endif
-
static void hmp_tcg_register(void)
{
monitor_register_hmp_info_hrt("jit", qmp_x_query_jit);
diff --git a/accel/tcg/tcg-accel-ops.c b/accel/tcg/tcg-accel-ops.c
index 58c8e64096..3973591508 100644
--- a/accel/tcg/tcg-accel-ops.c
+++ b/accel/tcg/tcg-accel-ops.c
@@ -70,20 +70,10 @@ void tcg_cpus_destroy(CPUState *cpu)
int tcg_cpus_exec(CPUState *cpu)
{
int ret;
-#ifdef CONFIG_PROFILER
- int64_t ti;
-#endif
assert(tcg_enabled());
-#ifdef CONFIG_PROFILER
- ti = profile_getclock();
-#endif
cpu_exec_start(cpu);
ret = cpu_exec(cpu);
cpu_exec_end(cpu);
-#ifdef CONFIG_PROFILER
- qatomic_set(&tcg_ctx->prof.cpu_exec_time,
- tcg_ctx->prof.cpu_exec_time + profile_getclock() - ti);
-#endif
return ret;
}
diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c
index 03c49baf1c..d3d4fbc1a4 100644
--- a/accel/tcg/translate-all.c
+++ b/accel/tcg/translate-all.c
@@ -202,10 +202,6 @@ void cpu_restore_state_from_tb(CPUState *cpu, TranslationBlock *tb,
uintptr_t host_pc)
{
uint64_t data[TARGET_INSN_START_WORDS];
-#ifdef CONFIG_PROFILER
- TCGProfile *prof = &tcg_ctx->prof;
- int64_t ti = profile_getclock();
-#endif
int insns_left = cpu_unwind_data_from_tb(tb, host_pc, data);
if (insns_left < 0) {
@@ -222,12 +218,6 @@ void cpu_restore_state_from_tb(CPUState *cpu, TranslationBlock *tb,
}
cpu->cc->tcg_ops->restore_state_to_opc(cpu, tb, data);
-
-#ifdef CONFIG_PROFILER
- qatomic_set(&prof->restore_time,
- prof->restore_time + profile_getclock() - ti);
- qatomic_set(&prof->restore_count, prof->restore_count + 1);
-#endif
}
bool cpu_restore_state(CPUState *cpu, uintptr_t host_pc)
@@ -290,13 +280,6 @@ static int setjmp_gen_code(CPUArchState *env, TranslationBlock *tb,
tcg_ctx->cpu = NULL;
*max_insns = tb->icount;
-#ifdef CONFIG_PROFILER
- qatomic_set(&tcg_ctx->prof.tb_count, tcg_ctx->prof.tb_count + 1);
- qatomic_set(&tcg_ctx->prof.interm_time,
- tcg_ctx->prof.interm_time + profile_getclock() - *ti);
- *ti = profile_getclock();
-#endif
-
return tcg_gen_code(tcg_ctx, tb, pc);
}
@@ -310,9 +293,6 @@ TranslationBlock *tb_gen_code(CPUState *cpu,
tb_page_addr_t phys_pc;
tcg_insn_unit *gen_code_buf;
int gen_code_size, search_size, max_insns;
-#ifdef CONFIG_PROFILER
- TCGProfile *prof = &tcg_ctx->prof;
-#endif
int64_t ti;
void *host_pc;
@@ -371,12 +351,6 @@ TranslationBlock *tb_gen_code(CPUState *cpu,
tb_overflow:
-#ifdef CONFIG_PROFILER
- /* includes aborted translations because of exceptions */
- qatomic_set(&prof->tb_count1, prof->tb_count1 + 1);
- ti = profile_getclock();
-#endif
-
trace_translate_block(tb, pc, tb->tc.ptr);
gen_code_size = setjmp_gen_code(env, tb, pc, host_pc, &max_insns, &ti);
@@ -431,13 +405,6 @@ TranslationBlock *tb_gen_code(CPUState *cpu,
*/
perf_report_code(pc, tb, tcg_splitwx_to_rx(gen_code_buf));
-#ifdef CONFIG_PROFILER
- qatomic_set(&prof->code_time, prof->code_time + profile_getclock() - ti);
- qatomic_set(&prof->code_in_len, prof->code_in_len + tb->size);
- qatomic_set(&prof->code_out_len, prof->code_out_len + gen_code_size);
- qatomic_set(&prof->search_out_len, prof->search_out_len + search_size);
-#endif
-
if (qemu_loglevel_mask(CPU_LOG_TB_OUT_ASM) &&
qemu_log_in_addr_range(pc)) {
FILE *logfile = qemu_log_trylock();
diff --git a/softmmu/runstate.c b/softmmu/runstate.c
index 1957caf73f..a9fbcf4862 100644
--- a/softmmu/runstate.c
+++ b/softmmu/runstate.c
@@ -727,18 +727,9 @@ static bool main_loop_should_exit(int *status)
int qemu_main_loop(void)
{
int status = EXIT_SUCCESS;
-#ifdef CONFIG_PROFILER
- int64_t ti;
-#endif
while (!main_loop_should_exit(&status)) {
-#ifdef CONFIG_PROFILER
- ti = profile_getclock();
-#endif
main_loop_wait(false);
-#ifdef CONFIG_PROFILER
- dev_time += profile_getclock() - ti;
-#endif
}
return status;
diff --git a/tcg/tcg.c b/tcg/tcg.c
index 3fcd0d9f32..a0628fe424 100644
--- a/tcg/tcg.c
+++ b/tcg/tcg.c
@@ -3033,10 +3033,6 @@ void tcg_op_remove(TCGContext *s, TCGOp *op)
QTAILQ_REMOVE(&s->ops, op, link);
QTAILQ_INSERT_TAIL(&s->free_ops, op, link);
s->nb_ops--;
-
-#ifdef CONFIG_PROFILER
- qatomic_set(&s->prof.del_op_count, s->prof.del_op_count + 1);
-#endif
}
void tcg_remove_ops_after(TCGOp *op)
@@ -5906,143 +5902,16 @@ static void tcg_out_st_helper_args(TCGContext *s, const TCGLabelQemuLdst *ldst,
tcg_out_helper_load_common_args(s, ldst, parm, info, next_arg);
}
-#ifdef CONFIG_PROFILER
-
-/* avoid copy/paste errors */
-#define PROF_ADD(to, from, field) \
- do { \
- (to)->field += qatomic_read(&((from)->field)); \
- } while (0)
-
-#define PROF_MAX(to, from, field) \
- do { \
- typeof((from)->field) val__ = qatomic_read(&((from)->field)); \
- if (val__ > (to)->field) { \
- (to)->field = val__; \
- } \
- } while (0)
-
-/* Pass in a zero'ed @prof */
-static inline
-void tcg_profile_snapshot(TCGProfile *prof, bool counters, bool table)
-{
- unsigned int n_ctxs = qatomic_read(&tcg_cur_ctxs);
- unsigned int i;
-
- for (i = 0; i < n_ctxs; i++) {
- TCGContext *s = qatomic_read(&tcg_ctxs[i]);
- const TCGProfile *orig = &s->prof;
-
- if (counters) {
- PROF_ADD(prof, orig, cpu_exec_time);
- PROF_ADD(prof, orig, tb_count1);
- PROF_ADD(prof, orig, tb_count);
- PROF_ADD(prof, orig, op_count);
- PROF_MAX(prof, orig, op_count_max);
- PROF_ADD(prof, orig, temp_count);
- PROF_MAX(prof, orig, temp_count_max);
- PROF_ADD(prof, orig, del_op_count);
- PROF_ADD(prof, orig, code_in_len);
- PROF_ADD(prof, orig, code_out_len);
- PROF_ADD(prof, orig, search_out_len);
- PROF_ADD(prof, orig, interm_time);
- PROF_ADD(prof, orig, code_time);
- PROF_ADD(prof, orig, la_time);
- PROF_ADD(prof, orig, opt_time);
- PROF_ADD(prof, orig, restore_count);
- PROF_ADD(prof, orig, restore_time);
- }
- if (table) {
- int i;
-
- for (i = 0; i < NB_OPS; i++) {
- PROF_ADD(prof, orig, table_op_count[i]);
- }
- }
- }
-}
-
-#undef PROF_ADD
-#undef PROF_MAX
-
-static void tcg_profile_snapshot_counters(TCGProfile *prof)
-{
- tcg_profile_snapshot(prof, true, false);
-}
-
-static void tcg_profile_snapshot_table(TCGProfile *prof)
-{
- tcg_profile_snapshot(prof, false, true);
-}
-
-void tcg_dump_op_count(GString *buf)
-{
- TCGProfile prof = {};
- int i;
-
- tcg_profile_snapshot_table(&prof);
- for (i = 0; i < NB_OPS; i++) {
- g_string_append_printf(buf, "%s %" PRId64 "\n", tcg_op_defs[i].name,
- prof.table_op_count[i]);
- }
-}
-
-int64_t tcg_cpu_exec_time(void)
-{
- unsigned int n_ctxs = qatomic_read(&tcg_cur_ctxs);
- unsigned int i;
- int64_t ret = 0;
-
- for (i = 0; i < n_ctxs; i++) {
- const TCGContext *s = qatomic_read(&tcg_ctxs[i]);
- const TCGProfile *prof = &s->prof;
-
- ret += qatomic_read(&prof->cpu_exec_time);
- }
- return ret;
-}
-#else
void tcg_dump_op_count(GString *buf)
{
g_string_append_printf(buf, "[TCG profiler not compiled]\n");
}
-int64_t tcg_cpu_exec_time(void)
-{
- error_report("%s: TCG profiler not compiled", __func__);
- exit(EXIT_FAILURE);
-}
-#endif
-
-
int tcg_gen_code(TCGContext *s, TranslationBlock *tb, uint64_t pc_start)
{
-#ifdef CONFIG_PROFILER
- TCGProfile *prof = &s->prof;
-#endif
int i, start_words, num_insns;
TCGOp *op;
-#ifdef CONFIG_PROFILER
- {
- int n = 0;
-
- QTAILQ_FOREACH(op, &s->ops, link) {
- n++;
- }
- qatomic_set(&prof->op_count, prof->op_count + n);
- if (n > prof->op_count_max) {
- qatomic_set(&prof->op_count_max, n);
- }
-
- n = s->nb_temps;
- qatomic_set(&prof->temp_count, prof->temp_count + n);
- if (n > prof->temp_count_max) {
- qatomic_set(&prof->temp_count_max, n);
- }
- }
-#endif
-
if (unlikely(qemu_loglevel_mask(CPU_LOG_TB_OP)
&& qemu_log_in_addr_range(pc_start))) {
FILE *logfile = qemu_log_trylock();
@@ -6071,17 +5940,8 @@ int tcg_gen_code(TCGContext *s, TranslationBlock *tb, uint64_t pc_start)
}
#endif
-#ifdef CONFIG_PROFILER
- qatomic_set(&prof->opt_time, prof->opt_time - profile_getclock());
-#endif
-
tcg_optimize(s);
-#ifdef CONFIG_PROFILER
- qatomic_set(&prof->opt_time, prof->opt_time + profile_getclock());
- qatomic_set(&prof->la_time, prof->la_time - profile_getclock());
-#endif
-
reachable_code_pass(s);
liveness_pass_0(s);
liveness_pass_1(s);
@@ -6105,10 +5965,6 @@ int tcg_gen_code(TCGContext *s, TranslationBlock *tb, uint64_t pc_start)
}
}
-#ifdef CONFIG_PROFILER
- qatomic_set(&prof->la_time, prof->la_time + profile_getclock());
-#endif
-
if (unlikely(qemu_loglevel_mask(CPU_LOG_TB_OP_OPT)
&& qemu_log_in_addr_range(pc_start))) {
FILE *logfile = qemu_log_trylock();
@@ -6151,10 +6007,6 @@ int tcg_gen_code(TCGContext *s, TranslationBlock *tb, uint64_t pc_start)
QTAILQ_FOREACH(op, &s->ops, link) {
TCGOpcode opc = op->opc;
-#ifdef CONFIG_PROFILER
- qatomic_set(&prof->table_op_count[opc], prof->table_op_count[opc] + 1);
-#endif
-
switch (opc) {
case INDEX_op_mov_i32:
case INDEX_op_mov_i64:
@@ -6249,76 +6101,10 @@ int tcg_gen_code(TCGContext *s, TranslationBlock *tb, uint64_t pc_start)
return tcg_current_code_size(s);
}
-#ifdef CONFIG_PROFILER
-void tcg_dump_info(GString *buf)
-{
- TCGProfile prof = {};
- const TCGProfile *s;
- int64_t tb_count;
- int64_t tb_div_count;
- int64_t tot;
-
- tcg_profile_snapshot_counters(&prof);
- s = &prof;
- tb_count = s->tb_count;
- tb_div_count = tb_count ? tb_count : 1;
- tot = s->interm_time + s->code_time;
-
- g_string_append_printf(buf, "JIT cycles %" PRId64
- " (%0.3f s at 2.4 GHz)\n",
- tot, tot / 2.4e9);
- g_string_append_printf(buf, "translated TBs %" PRId64
- " (aborted=%" PRId64 " %0.1f%%)\n",
- tb_count, s->tb_count1 - tb_count,
- (double)(s->tb_count1 - s->tb_count)
- / (s->tb_count1 ? s->tb_count1 : 1) * 100.0);
- g_string_append_printf(buf, "avg ops/TB %0.1f max=%d\n",
- (double)s->op_count / tb_div_count, s->op_count_max);
- g_string_append_printf(buf, "deleted ops/TB %0.2f\n",
- (double)s->del_op_count / tb_div_count);
- g_string_append_printf(buf, "avg temps/TB %0.2f max=%d\n",
- (double)s->temp_count / tb_div_count,
- s->temp_count_max);
- g_string_append_printf(buf, "avg host code/TB %0.1f\n",
- (double)s->code_out_len / tb_div_count);
- g_string_append_printf(buf, "avg search data/TB %0.1f\n",
- (double)s->search_out_len / tb_div_count);
-
- g_string_append_printf(buf, "cycles/op %0.1f\n",
- s->op_count ? (double)tot / s->op_count : 0);
- g_string_append_printf(buf, "cycles/in byte %0.1f\n",
- s->code_in_len ? (double)tot / s->code_in_len : 0);
- g_string_append_printf(buf, "cycles/out byte %0.1f\n",
- s->code_out_len ? (double)tot / s->code_out_len : 0);
- g_string_append_printf(buf, "cycles/search byte %0.1f\n",
- s->search_out_len ?
- (double)tot / s->search_out_len : 0);
- if (tot == 0) {
- tot = 1;
- }
- g_string_append_printf(buf, " gen_interm time %0.1f%%\n",
- (double)s->interm_time / tot * 100.0);
- g_string_append_printf(buf, " gen_code time %0.1f%%\n",
- (double)s->code_time / tot * 100.0);
- g_string_append_printf(buf, "optim./code time %0.1f%%\n",
- (double)s->opt_time / (s->code_time ?
- s->code_time : 1)
- * 100.0);
- g_string_append_printf(buf, "liveness/code time %0.1f%%\n",
- (double)s->la_time / (s->code_time ?
- s->code_time : 1) * 100.0);
- g_string_append_printf(buf, "cpu_restore count %" PRId64 "\n",
- s->restore_count);
- g_string_append_printf(buf, " avg cycles %0.1f\n",
- s->restore_count ?
- (double)s->restore_time / s->restore_count : 0);
-}
-#else
void tcg_dump_info(GString *buf)
{
g_string_append_printf(buf, "[TCG profiler not compiled]\n");
}
-#endif
#ifdef ELF_HOST_MACHINE
/* In order to use this feature, the backend needs to do three things:
diff --git a/tests/qtest/qmp-cmd-test.c b/tests/qtest/qmp-cmd-test.c
index a58de48d2a..73a670e8fa 100644
--- a/tests/qtest/qmp-cmd-test.c
+++ b/tests/qtest/qmp-cmd-test.c
@@ -46,9 +46,6 @@ static int query_error_class(const char *cmd)
{ "query-balloon", ERROR_CLASS_DEVICE_NOT_ACTIVE },
{ "query-hotpluggable-cpus", ERROR_CLASS_GENERIC_ERROR },
{ "query-vm-generation-id", ERROR_CLASS_GENERIC_ERROR },
-#ifndef CONFIG_PROFILER
- { "x-query-profile", ERROR_CLASS_GENERIC_ERROR },
-#endif
/* Only valid with a USB bus added */
{ "x-query-usb", ERROR_CLASS_GENERIC_ERROR },
/* Only valid with accel=tcg */
diff --git a/hmp-commands-info.hx b/hmp-commands-info.hx
index 47d63d26db..f5b37eb74a 100644
--- a/hmp-commands-info.hx
+++ b/hmp-commands-info.hx
@@ -360,21 +360,6 @@ SRST
Show host USB devices.
ERST
-#if defined(CONFIG_TCG)
- {
- .name = "profile",
- .args_type = "",
- .params = "",
- .help = "show profiling information",
- .cmd_info_hrt = qmp_x_query_profile,
- },
-#endif
-
-SRST
- ``info profile``
- Show profiling information.
-ERST
-
{
.name = "capture",
.args_type = "",
diff --git a/meson_options.txt b/meson_options.txt
index 90237389e2..bbb5c7e886 100644
--- a/meson_options.txt
+++ b/meson_options.txt
@@ -345,8 +345,6 @@ option('qom_cast_debug', type: 'boolean', value: true,
option('gprof', type: 'boolean', value: false,
description: 'QEMU profiling with gprof',
deprecated: true)
-option('profiler', type: 'boolean', value: false,
- description: 'profiler support')
option('slirp_smbd', type : 'feature', value : 'auto',
description: 'use smbd (at path --smbd=*) in slirp networking')
diff --git a/scripts/meson-buildoptions.sh b/scripts/meson-buildoptions.sh
index 5714fd93d9..7dd5709ef4 100644
--- a/scripts/meson-buildoptions.sh
+++ b/scripts/meson-buildoptions.sh
@@ -39,7 +39,6 @@ meson_options_help() {
printf "%s\n" ' jemalloc/system/tcmalloc)'
printf "%s\n" ' --enable-module-upgrades try to load modules from alternate paths for'
printf "%s\n" ' upgrades'
- printf "%s\n" ' --enable-profiler profiler support'
printf "%s\n" ' --enable-rng-none dummy RNG, avoid using /dev/(u)random and'
printf "%s\n" ' getrandom()'
printf "%s\n" ' --enable-safe-stack SafeStack Stack Smash Protection (requires'
@@ -401,8 +400,6 @@ _meson_option_parse() {
--with-pkgversion=*) quote_sh "-Dpkgversion=$2" ;;
--enable-png) printf "%s" -Dpng=enabled ;;
--disable-png) printf "%s" -Dpng=disabled ;;
- --enable-profiler) printf "%s" -Dprofiler=true ;;
- --disable-profiler) printf "%s" -Dprofiler=false ;;
--enable-pvrdma) printf "%s" -Dpvrdma=enabled ;;
--disable-pvrdma) printf "%s" -Dpvrdma=disabled ;;
--enable-qcow1) printf "%s" -Dqcow1=enabled ;;
--
2.34.1
^ permalink raw reply related [flat|nested] 28+ messages in thread* [PULL 15/22] tcg: Fix temporary variable in tcg_gen_gvec_andcs
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
` (13 preceding siblings ...)
2023-06-26 15:39 ` [PULL 14/22] accel/tcg: remove CONFIG_PROFILER Richard Henderson
@ 2023-06-26 15:39 ` Richard Henderson
2023-06-26 15:39 ` [PULL 16/22] target/microblaze: Define TCG_GUEST_DEFAULT_MO Richard Henderson
` (7 subsequent siblings)
22 siblings, 0 replies; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 15:39 UTC (permalink / raw)
To: qemu-devel; +Cc: Max Chou, Daniel Henrique Barboza
From: Max Chou <max.chou@sifive.com>
The 5th parameter of tcg_gen_gvec_2s should be replaced by the
temporary tmp variable in the tcg_gen_gvec_andcs function.
Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Signed-off-by: Max Chou <max.chou@sifive.com>
Message-Id: <20230622161646.32005-9-max.chou@sifive.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
tcg/tcg-op-gvec.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tcg/tcg-op-gvec.c b/tcg/tcg-op-gvec.c
index 95a588d6d2..a062239804 100644
--- a/tcg/tcg-op-gvec.c
+++ b/tcg/tcg-op-gvec.c
@@ -2774,7 +2774,7 @@ void tcg_gen_gvec_andcs(unsigned vece, uint32_t dofs, uint32_t aofs,
TCGv_i64 tmp = tcg_temp_ebb_new_i64();
tcg_gen_dup_i64(vece, tmp, c);
- tcg_gen_gvec_2s(dofs, aofs, oprsz, maxsz, c, &g);
+ tcg_gen_gvec_2s(dofs, aofs, oprsz, maxsz, tmp, &g);
tcg_temp_free_i64(tmp);
}
--
2.34.1
^ permalink raw reply related [flat|nested] 28+ messages in thread* [PULL 16/22] target/microblaze: Define TCG_GUEST_DEFAULT_MO
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
` (14 preceding siblings ...)
2023-06-26 15:39 ` [PULL 15/22] tcg: Fix temporary variable in tcg_gen_gvec_andcs Richard Henderson
@ 2023-06-26 15:39 ` Richard Henderson
2023-06-26 15:39 ` [PULL 17/22] tcg: Do not elide memory barriers for !CF_PARALLEL in system mode Richard Henderson
` (6 subsequent siblings)
22 siblings, 0 replies; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 15:39 UTC (permalink / raw)
To: qemu-devel; +Cc: Philippe Mathieu-Daudé, Edgar E . Iglesias
The microblaze architecture does not reorder instructions.
While there is an MBAR wait-for-data-access instruction,
this concerns synchronizing with DMA.
This should have been defined when enabling MTTCG.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar@zeroasic.com>
Fixes: d449561b130 ("configure: microblaze: Enable mttcg")
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
target/microblaze/cpu.h | 3 +++
1 file changed, 3 insertions(+)
diff --git a/target/microblaze/cpu.h b/target/microblaze/cpu.h
index 3525de144c..a7b040abd4 100644
--- a/target/microblaze/cpu.h
+++ b/target/microblaze/cpu.h
@@ -24,6 +24,9 @@
#include "exec/cpu-defs.h"
#include "qemu/cpu-float.h"
+/* MicroBlaze is always in-order. */
+#define TCG_GUEST_DEFAULT_MO TCG_MO_ALL
+
typedef struct CPUArchState CPUMBState;
#if !defined(CONFIG_USER_ONLY)
#include "mmu.h"
--
2.34.1
^ permalink raw reply related [flat|nested] 28+ messages in thread* [PULL 17/22] tcg: Do not elide memory barriers for !CF_PARALLEL in system mode
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
` (15 preceding siblings ...)
2023-06-26 15:39 ` [PULL 16/22] target/microblaze: Define TCG_GUEST_DEFAULT_MO Richard Henderson
@ 2023-06-26 15:39 ` Richard Henderson
2023-06-26 15:39 ` [PULL 18/22] tcg: Add host memory barriers to cpu_ldst.h interfaces Richard Henderson
` (5 subsequent siblings)
22 siblings, 0 replies; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 15:39 UTC (permalink / raw)
To: qemu-devel; +Cc: Philippe Mathieu-Daudé
The virtio devices require proper memory ordering between
the vcpus and the iothreads.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
tcg/tcg-op.c | 14 +++++++++++++-
1 file changed, 13 insertions(+), 1 deletion(-)
diff --git a/tcg/tcg-op.c b/tcg/tcg-op.c
index c07de5d9f8..7aadb37756 100644
--- a/tcg/tcg-op.c
+++ b/tcg/tcg-op.c
@@ -102,7 +102,19 @@ void tcg_gen_br(TCGLabel *l)
void tcg_gen_mb(TCGBar mb_type)
{
- if (tcg_ctx->gen_tb->cflags & CF_PARALLEL) {
+#ifdef CONFIG_USER_ONLY
+ bool parallel = tcg_ctx->gen_tb->cflags & CF_PARALLEL;
+#else
+ /*
+ * It is tempting to elide the barrier in a uniprocessor context.
+ * However, even with a single cpu we have i/o threads running in
+ * parallel, and lack of memory order can result in e.g. virtio
+ * queue entries being read incorrectly.
+ */
+ bool parallel = true;
+#endif
+
+ if (parallel) {
tcg_gen_op1(INDEX_op_mb, mb_type);
}
}
--
2.34.1
^ permalink raw reply related [flat|nested] 28+ messages in thread* [PULL 18/22] tcg: Add host memory barriers to cpu_ldst.h interfaces
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
` (16 preceding siblings ...)
2023-06-26 15:39 ` [PULL 17/22] tcg: Do not elide memory barriers for !CF_PARALLEL in system mode Richard Henderson
@ 2023-06-26 15:39 ` Richard Henderson
2023-06-26 15:39 ` [PULL 19/22] accel/tcg: Remove check_tcg_memory_orders_compatible Richard Henderson
` (4 subsequent siblings)
22 siblings, 0 replies; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 15:39 UTC (permalink / raw)
To: qemu-devel; +Cc: Philippe Mathieu-Daudé
Bring the helpers into line with the rest of tcg in respecting
guest memory ordering.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
accel/tcg/internal.h | 34 ++++++++++++++++++++++++++++++++++
accel/tcg/cputlb.c | 10 ++++++++++
accel/tcg/user-exec.c | 10 ++++++++++
3 files changed, 54 insertions(+)
diff --git a/accel/tcg/internal.h b/accel/tcg/internal.h
index 91f308bdfa..650c3ac53f 100644
--- a/accel/tcg/internal.h
+++ b/accel/tcg/internal.h
@@ -78,4 +78,38 @@ extern int64_t max_advance;
extern bool one_insn_per_tb;
+/**
+ * tcg_req_mo:
+ * @type: TCGBar
+ *
+ * Filter @type to the barrier that is required for the guest
+ * memory ordering vs the host memory ordering. A non-zero
+ * result indicates that some barrier is required.
+ *
+ * If TCG_GUEST_DEFAULT_MO is not defined, assume that the
+ * guest requires strict ordering.
+ *
+ * This is a macro so that it's constant even without optimization.
+ */
+#ifdef TCG_GUEST_DEFAULT_MO
+# define tcg_req_mo(type) \
+ ((type) & TCG_GUEST_DEFAULT_MO & ~TCG_TARGET_DEFAULT_MO)
+#else
+# define tcg_req_mo(type) ((type) & ~TCG_TARGET_DEFAULT_MO)
+#endif
+
+/**
+ * cpu_req_mo:
+ * @type: TCGBar
+ *
+ * If tcg_req_mo indicates a barrier for @type is required
+ * for the guest memory model, issue a host memory barrier.
+ */
+#define cpu_req_mo(type) \
+ do { \
+ if (tcg_req_mo(type)) { \
+ smp_mb(); \
+ } \
+ } while (0)
+
#endif /* ACCEL_TCG_INTERNAL_H */
diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c
index e02cfc550e..5666a8e23a 100644
--- a/accel/tcg/cputlb.c
+++ b/accel/tcg/cputlb.c
@@ -2339,6 +2339,7 @@ static uint8_t do_ld1_mmu(CPUArchState *env, vaddr addr, MemOpIdx oi,
MMULookupLocals l;
bool crosspage;
+ cpu_req_mo(TCG_MO_LD_LD | TCG_MO_ST_LD);
crosspage = mmu_lookup(env, addr, oi, ra, access_type, &l);
tcg_debug_assert(!crosspage);
@@ -2360,6 +2361,7 @@ static uint16_t do_ld2_mmu(CPUArchState *env, vaddr addr, MemOpIdx oi,
uint16_t ret;
uint8_t a, b;
+ cpu_req_mo(TCG_MO_LD_LD | TCG_MO_ST_LD);
crosspage = mmu_lookup(env, addr, oi, ra, access_type, &l);
if (likely(!crosspage)) {
return do_ld_2(env, &l.page[0], l.mmu_idx, access_type, l.memop, ra);
@@ -2390,6 +2392,7 @@ static uint32_t do_ld4_mmu(CPUArchState *env, vaddr addr, MemOpIdx oi,
bool crosspage;
uint32_t ret;
+ cpu_req_mo(TCG_MO_LD_LD | TCG_MO_ST_LD);
crosspage = mmu_lookup(env, addr, oi, ra, access_type, &l);
if (likely(!crosspage)) {
return do_ld_4(env, &l.page[0], l.mmu_idx, access_type, l.memop, ra);
@@ -2417,6 +2420,7 @@ static uint64_t do_ld8_mmu(CPUArchState *env, vaddr addr, MemOpIdx oi,
bool crosspage;
uint64_t ret;
+ cpu_req_mo(TCG_MO_LD_LD | TCG_MO_ST_LD);
crosspage = mmu_lookup(env, addr, oi, ra, access_type, &l);
if (likely(!crosspage)) {
return do_ld_8(env, &l.page[0], l.mmu_idx, access_type, l.memop, ra);
@@ -2469,6 +2473,7 @@ static Int128 do_ld16_mmu(CPUArchState *env, vaddr addr,
Int128 ret;
int first;
+ cpu_req_mo(TCG_MO_LD_LD | TCG_MO_ST_LD);
crosspage = mmu_lookup(env, addr, oi, ra, MMU_DATA_LOAD, &l);
if (likely(!crosspage)) {
/* Perform the load host endian. */
@@ -2802,6 +2807,7 @@ void helper_stb_mmu(CPUArchState *env, uint64_t addr, uint32_t val,
bool crosspage;
tcg_debug_assert((get_memop(oi) & MO_SIZE) == MO_8);
+ cpu_req_mo(TCG_MO_LD_ST | TCG_MO_ST_ST);
crosspage = mmu_lookup(env, addr, oi, ra, MMU_DATA_STORE, &l);
tcg_debug_assert(!crosspage);
@@ -2815,6 +2821,7 @@ static void do_st2_mmu(CPUArchState *env, vaddr addr, uint16_t val,
bool crosspage;
uint8_t a, b;
+ cpu_req_mo(TCG_MO_LD_ST | TCG_MO_ST_ST);
crosspage = mmu_lookup(env, addr, oi, ra, MMU_DATA_STORE, &l);
if (likely(!crosspage)) {
do_st_2(env, &l.page[0], val, l.mmu_idx, l.memop, ra);
@@ -2843,6 +2850,7 @@ static void do_st4_mmu(CPUArchState *env, vaddr addr, uint32_t val,
MMULookupLocals l;
bool crosspage;
+ cpu_req_mo(TCG_MO_LD_ST | TCG_MO_ST_ST);
crosspage = mmu_lookup(env, addr, oi, ra, MMU_DATA_STORE, &l);
if (likely(!crosspage)) {
do_st_4(env, &l.page[0], val, l.mmu_idx, l.memop, ra);
@@ -2870,6 +2878,7 @@ static void do_st8_mmu(CPUArchState *env, vaddr addr, uint64_t val,
MMULookupLocals l;
bool crosspage;
+ cpu_req_mo(TCG_MO_LD_ST | TCG_MO_ST_ST);
crosspage = mmu_lookup(env, addr, oi, ra, MMU_DATA_STORE, &l);
if (likely(!crosspage)) {
do_st_8(env, &l.page[0], val, l.mmu_idx, l.memop, ra);
@@ -2899,6 +2908,7 @@ static void do_st16_mmu(CPUArchState *env, vaddr addr, Int128 val,
uint64_t a, b;
int first;
+ cpu_req_mo(TCG_MO_LD_ST | TCG_MO_ST_ST);
crosspage = mmu_lookup(env, addr, oi, ra, MMU_DATA_STORE, &l);
if (likely(!crosspage)) {
/* Swap to host endian if necessary, then store. */
diff --git a/accel/tcg/user-exec.c b/accel/tcg/user-exec.c
index f8b16d6ab8..8fbcbf9771 100644
--- a/accel/tcg/user-exec.c
+++ b/accel/tcg/user-exec.c
@@ -914,6 +914,7 @@ static uint8_t do_ld1_mmu(CPUArchState *env, abi_ptr addr,
uint8_t ret;
tcg_debug_assert((mop & MO_SIZE) == MO_8);
+ cpu_req_mo(TCG_MO_LD_LD | TCG_MO_ST_LD);
haddr = cpu_mmu_lookup(env, addr, mop, ra, MMU_DATA_LOAD);
ret = ldub_p(haddr);
clear_helper_retaddr();
@@ -947,6 +948,7 @@ static uint16_t do_ld2_mmu(CPUArchState *env, abi_ptr addr,
uint16_t ret;
tcg_debug_assert((mop & MO_SIZE) == MO_16);
+ cpu_req_mo(TCG_MO_LD_LD | TCG_MO_ST_LD);
haddr = cpu_mmu_lookup(env, addr, mop, ra, MMU_DATA_LOAD);
ret = load_atom_2(env, ra, haddr, mop);
clear_helper_retaddr();
@@ -984,6 +986,7 @@ static uint32_t do_ld4_mmu(CPUArchState *env, abi_ptr addr,
uint32_t ret;
tcg_debug_assert((mop & MO_SIZE) == MO_32);
+ cpu_req_mo(TCG_MO_LD_LD | TCG_MO_ST_LD);
haddr = cpu_mmu_lookup(env, addr, mop, ra, MMU_DATA_LOAD);
ret = load_atom_4(env, ra, haddr, mop);
clear_helper_retaddr();
@@ -1021,6 +1024,7 @@ static uint64_t do_ld8_mmu(CPUArchState *env, abi_ptr addr,
uint64_t ret;
tcg_debug_assert((mop & MO_SIZE) == MO_64);
+ cpu_req_mo(TCG_MO_LD_LD | TCG_MO_ST_LD);
haddr = cpu_mmu_lookup(env, addr, mop, ra, MMU_DATA_LOAD);
ret = load_atom_8(env, ra, haddr, mop);
clear_helper_retaddr();
@@ -1052,6 +1056,7 @@ static Int128 do_ld16_mmu(CPUArchState *env, abi_ptr addr,
Int128 ret;
tcg_debug_assert((mop & MO_SIZE) == MO_128);
+ cpu_req_mo(TCG_MO_LD_LD | TCG_MO_ST_LD);
haddr = cpu_mmu_lookup(env, addr, mop, ra, MMU_DATA_LOAD);
ret = load_atom_16(env, ra, haddr, mop);
clear_helper_retaddr();
@@ -1087,6 +1092,7 @@ static void do_st1_mmu(CPUArchState *env, abi_ptr addr, uint8_t val,
void *haddr;
tcg_debug_assert((mop & MO_SIZE) == MO_8);
+ cpu_req_mo(TCG_MO_LD_ST | TCG_MO_ST_ST);
haddr = cpu_mmu_lookup(env, addr, mop, ra, MMU_DATA_STORE);
stb_p(haddr, val);
clear_helper_retaddr();
@@ -1111,6 +1117,7 @@ static void do_st2_mmu(CPUArchState *env, abi_ptr addr, uint16_t val,
void *haddr;
tcg_debug_assert((mop & MO_SIZE) == MO_16);
+ cpu_req_mo(TCG_MO_LD_ST | TCG_MO_ST_ST);
haddr = cpu_mmu_lookup(env, addr, mop, ra, MMU_DATA_STORE);
if (mop & MO_BSWAP) {
@@ -1139,6 +1146,7 @@ static void do_st4_mmu(CPUArchState *env, abi_ptr addr, uint32_t val,
void *haddr;
tcg_debug_assert((mop & MO_SIZE) == MO_32);
+ cpu_req_mo(TCG_MO_LD_ST | TCG_MO_ST_ST);
haddr = cpu_mmu_lookup(env, addr, mop, ra, MMU_DATA_STORE);
if (mop & MO_BSWAP) {
@@ -1167,6 +1175,7 @@ static void do_st8_mmu(CPUArchState *env, abi_ptr addr, uint64_t val,
void *haddr;
tcg_debug_assert((mop & MO_SIZE) == MO_64);
+ cpu_req_mo(TCG_MO_LD_ST | TCG_MO_ST_ST);
haddr = cpu_mmu_lookup(env, addr, mop, ra, MMU_DATA_STORE);
if (mop & MO_BSWAP) {
@@ -1195,6 +1204,7 @@ static void do_st16_mmu(CPUArchState *env, abi_ptr addr, Int128 val,
void *haddr;
tcg_debug_assert((mop & MO_SIZE) == MO_128);
+ cpu_req_mo(TCG_MO_LD_ST | TCG_MO_ST_ST);
haddr = cpu_mmu_lookup(env, addr, mop, ra, MMU_DATA_STORE);
if (mop & MO_BSWAP) {
--
2.34.1
^ permalink raw reply related [flat|nested] 28+ messages in thread* [PULL 19/22] accel/tcg: Remove check_tcg_memory_orders_compatible
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
` (17 preceding siblings ...)
2023-06-26 15:39 ` [PULL 18/22] tcg: Add host memory barriers to cpu_ldst.h interfaces Richard Henderson
@ 2023-06-26 15:39 ` Richard Henderson
2023-06-26 15:39 ` [PULL 20/22] accel/tcg: Store some tlb flags in CPUTLBEntryFull Richard Henderson
` (3 subsequent siblings)
22 siblings, 0 replies; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 15:39 UTC (permalink / raw)
To: qemu-devel; +Cc: Philippe Mathieu-Daudé
We now issue host memory barriers to match the guest memory order.
Continue to disable MTTCG only if the guest has not been ported.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
accel/tcg/tcg-all.c | 39 ++++++++++-----------------------------
1 file changed, 10 insertions(+), 29 deletions(-)
diff --git a/accel/tcg/tcg-all.c b/accel/tcg/tcg-all.c
index 02af6a2891..03dfd67e9e 100644
--- a/accel/tcg/tcg-all.c
+++ b/accel/tcg/tcg-all.c
@@ -64,37 +64,23 @@ DECLARE_INSTANCE_CHECKER(TCGState, TCG_STATE,
* they can set the appropriate CONFIG flags in ${target}-softmmu.mak
*
* Once a guest architecture has been converted to the new primitives
- * there are two remaining limitations to check.
- *
- * - The guest can't be oversized (e.g. 64 bit guest on 32 bit host)
- * - The host must have a stronger memory order than the guest
- *
- * It may be possible in future to support strong guests on weak hosts
- * but that will require tagging all load/stores in a guest with their
- * implicit memory order requirements which would likely slow things
- * down a lot.
+ * there is one remaining limitation to check:
+ * - The guest can't be oversized (e.g. 64 bit guest on 32 bit host)
*/
-static bool check_tcg_memory_orders_compatible(void)
-{
-#if defined(TCG_GUEST_DEFAULT_MO) && defined(TCG_TARGET_DEFAULT_MO)
- return (TCG_GUEST_DEFAULT_MO & ~TCG_TARGET_DEFAULT_MO) == 0;
-#else
- return false;
-#endif
-}
-
static bool default_mttcg_enabled(void)
{
if (icount_enabled() || TCG_OVERSIZED_GUEST) {
return false;
- } else {
-#ifdef TARGET_SUPPORTS_MTTCG
- return check_tcg_memory_orders_compatible();
-#else
- return false;
-#endif
}
+#ifdef TARGET_SUPPORTS_MTTCG
+# ifndef TCG_GUEST_DEFAULT_MO
+# error "TARGET_SUPPORTS_MTTCG without TCG_GUEST_DEFAULT_MO"
+# endif
+ return true;
+#else
+ return false;
+#endif
}
static void tcg_accel_instance_init(Object *obj)
@@ -162,11 +148,6 @@ static void tcg_set_thread(Object *obj, const char *value, Error **errp)
warn_report("Guest not yet converted to MTTCG - "
"you may get unexpected results");
#endif
- if (!check_tcg_memory_orders_compatible()) {
- warn_report("Guest expects a stronger memory ordering "
- "than the host provides");
- error_printf("This may cause strange/hard to debug errors\n");
- }
s->mttcg_enabled = true;
}
} else if (strcmp(value, "single") == 0) {
--
2.34.1
^ permalink raw reply related [flat|nested] 28+ messages in thread* [PULL 20/22] accel/tcg: Store some tlb flags in CPUTLBEntryFull
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
` (18 preceding siblings ...)
2023-06-26 15:39 ` [PULL 19/22] accel/tcg: Remove check_tcg_memory_orders_compatible Richard Henderson
@ 2023-06-26 15:39 ` Richard Henderson
2023-06-26 15:39 ` [PULL 21/22] accel/tcg: Move TLB_WATCHPOINT to TLB_SLOW_FLAGS_MASK Richard Henderson
` (2 subsequent siblings)
22 siblings, 0 replies; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 15:39 UTC (permalink / raw)
To: qemu-devel; +Cc: Philippe Mathieu-Daudé
We have run out of bits we can use within the CPUTLBEntry comparators,
as TLB_FLAGS_MASK cannot overlap alignment.
Store slow_flags[] in CPUTLBEntryFull, and merge with the flags from
the comparator. A new TLB_FORCE_SLOW bit is set within the comparator
as an indication that the slow path must be used.
Move TLB_BSWAP to TLB_SLOW_FLAGS_MASK. Since we are out of bits,
we cannot create a new bit without moving an old one.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
include/exec/cpu-all.h | 21 +++++++--
include/exec/cpu-defs.h | 6 +++
include/hw/core/cpu.h | 1 +
accel/tcg/cputlb.c | 98 ++++++++++++++++++++++++-----------------
4 files changed, 82 insertions(+), 44 deletions(-)
diff --git a/include/exec/cpu-all.h b/include/exec/cpu-all.h
index 09bf4c0cc6..4422f4bb07 100644
--- a/include/exec/cpu-all.h
+++ b/include/exec/cpu-all.h
@@ -327,17 +327,30 @@ CPUArchState *cpu_copy(CPUArchState *env);
#define TLB_MMIO (1 << (TARGET_PAGE_BITS_MIN - 3))
/* Set if TLB entry contains a watchpoint. */
#define TLB_WATCHPOINT (1 << (TARGET_PAGE_BITS_MIN - 4))
-/* Set if TLB entry requires byte swap. */
-#define TLB_BSWAP (1 << (TARGET_PAGE_BITS_MIN - 5))
+/* Set if the slow path must be used; more flags in CPUTLBEntryFull. */
+#define TLB_FORCE_SLOW (1 << (TARGET_PAGE_BITS_MIN - 5))
/* Set if TLB entry writes ignored. */
#define TLB_DISCARD_WRITE (1 << (TARGET_PAGE_BITS_MIN - 6))
-/* Use this mask to check interception with an alignment mask
+/*
+ * Use this mask to check interception with an alignment mask
* in a TCG backend.
*/
#define TLB_FLAGS_MASK \
(TLB_INVALID_MASK | TLB_NOTDIRTY | TLB_MMIO \
- | TLB_WATCHPOINT | TLB_BSWAP | TLB_DISCARD_WRITE)
+ | TLB_WATCHPOINT | TLB_FORCE_SLOW | TLB_DISCARD_WRITE)
+
+/*
+ * Flags stored in CPUTLBEntryFull.slow_flags[x].
+ * TLB_FORCE_SLOW must be set in CPUTLBEntry.addr_idx[x].
+ */
+/* Set if TLB entry requires byte swap. */
+#define TLB_BSWAP (1 << 0)
+
+#define TLB_SLOW_FLAGS_MASK TLB_BSWAP
+
+/* The two sets of flags must not overlap. */
+QEMU_BUILD_BUG_ON(TLB_FLAGS_MASK & TLB_SLOW_FLAGS_MASK);
/**
* tlb_hit_page: return true if page aligned @addr is a hit against the
diff --git a/include/exec/cpu-defs.h b/include/exec/cpu-defs.h
index e6a079402e..fb4c8d480f 100644
--- a/include/exec/cpu-defs.h
+++ b/include/exec/cpu-defs.h
@@ -124,6 +124,12 @@ typedef struct CPUTLBEntryFull {
/* @lg_page_size contains the log2 of the page size. */
uint8_t lg_page_size;
+ /*
+ * Additional tlb flags for use by the slow path. If non-zero,
+ * the corresponding CPUTLBEntry comparator must have TLB_FORCE_SLOW.
+ */
+ uint8_t slow_flags[MMU_ACCESS_COUNT];
+
/*
* Allow target-specific additions to this structure.
* This may be used to cache items from the guest cpu
diff --git a/include/hw/core/cpu.h b/include/hw/core/cpu.h
index 3b765beb9b..eda0230a02 100644
--- a/include/hw/core/cpu.h
+++ b/include/hw/core/cpu.h
@@ -84,6 +84,7 @@ typedef enum MMUAccessType {
MMU_DATA_LOAD = 0,
MMU_DATA_STORE = 1,
MMU_INST_FETCH = 2
+#define MMU_ACCESS_COUNT 3
} MMUAccessType;
typedef struct CPUWatchpoint CPUWatchpoint;
diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c
index 5666a8e23a..3671846744 100644
--- a/accel/tcg/cputlb.c
+++ b/accel/tcg/cputlb.c
@@ -1107,6 +1107,24 @@ static void tlb_add_large_page(CPUArchState *env, int mmu_idx,
env_tlb(env)->d[mmu_idx].large_page_mask = lp_mask;
}
+static inline void tlb_set_compare(CPUTLBEntryFull *full, CPUTLBEntry *ent,
+ target_ulong address, int flags,
+ MMUAccessType access_type, bool enable)
+{
+ if (enable) {
+ address |= flags & TLB_FLAGS_MASK;
+ flags &= TLB_SLOW_FLAGS_MASK;
+ if (flags) {
+ address |= TLB_FORCE_SLOW;
+ }
+ } else {
+ address = -1;
+ flags = 0;
+ }
+ ent->addr_idx[access_type] = address;
+ full->slow_flags[access_type] = flags;
+}
+
/*
* Add a new TLB entry. At most one entry for a given virtual address
* is permitted. Only a single TARGET_PAGE_SIZE region is mapped, the
@@ -1122,9 +1140,7 @@ void tlb_set_page_full(CPUState *cpu, int mmu_idx,
CPUTLB *tlb = env_tlb(env);
CPUTLBDesc *desc = &tlb->d[mmu_idx];
MemoryRegionSection *section;
- unsigned int index;
- vaddr address;
- vaddr write_address;
+ unsigned int index, read_flags, write_flags;
uintptr_t addend;
CPUTLBEntry *te, tn;
hwaddr iotlb, xlat, sz, paddr_page;
@@ -1153,13 +1169,13 @@ void tlb_set_page_full(CPUState *cpu, int mmu_idx,
" prot=%x idx=%d\n",
addr, full->phys_addr, prot, mmu_idx);
- address = addr_page;
+ read_flags = 0;
if (full->lg_page_size < TARGET_PAGE_BITS) {
/* Repeat the MMU check and TLB fill on every access. */
- address |= TLB_INVALID_MASK;
+ read_flags |= TLB_INVALID_MASK;
}
if (full->attrs.byte_swap) {
- address |= TLB_BSWAP;
+ read_flags |= TLB_BSWAP;
}
is_ram = memory_region_is_ram(section->mr);
@@ -1173,7 +1189,7 @@ void tlb_set_page_full(CPUState *cpu, int mmu_idx,
addend = 0;
}
- write_address = address;
+ write_flags = read_flags;
if (is_ram) {
iotlb = memory_region_get_ram_addr(section->mr) + xlat;
/*
@@ -1182,9 +1198,9 @@ void tlb_set_page_full(CPUState *cpu, int mmu_idx,
*/
if (prot & PAGE_WRITE) {
if (section->readonly) {
- write_address |= TLB_DISCARD_WRITE;
+ write_flags |= TLB_DISCARD_WRITE;
} else if (cpu_physical_memory_is_clean(iotlb)) {
- write_address |= TLB_NOTDIRTY;
+ write_flags |= TLB_NOTDIRTY;
}
}
} else {
@@ -1195,9 +1211,9 @@ void tlb_set_page_full(CPUState *cpu, int mmu_idx,
* Reads to romd devices go through the ram_ptr found above,
* but of course reads to I/O must go through MMIO.
*/
- write_address |= TLB_MMIO;
+ write_flags |= TLB_MMIO;
if (!is_romd) {
- address = write_address;
+ read_flags = write_flags;
}
}
@@ -1242,7 +1258,7 @@ void tlb_set_page_full(CPUState *cpu, int mmu_idx,
* TARGET_PAGE_BITS, and either
* + the ram_addr_t of the page base of the target RAM (RAM)
* + the offset within section->mr of the page base (I/O, ROMD)
- * We subtract the vaddr_page (which is page aligned and thus won't
+ * We subtract addr_page (which is page aligned and thus won't
* disturb the low bits) to give an offset which can be added to the
* (non-page-aligned) vaddr of the eventual memory access to get
* the MemoryRegion offset for the access. Note that the vaddr we
@@ -1250,36 +1266,30 @@ void tlb_set_page_full(CPUState *cpu, int mmu_idx,
* vaddr we add back in io_readx()/io_writex()/get_page_addr_code().
*/
desc->fulltlb[index] = *full;
- desc->fulltlb[index].xlat_section = iotlb - addr_page;
- desc->fulltlb[index].phys_addr = paddr_page;
+ full = &desc->fulltlb[index];
+ full->xlat_section = iotlb - addr_page;
+ full->phys_addr = paddr_page;
/* Now calculate the new entry */
tn.addend = addend - addr_page;
- if (prot & PAGE_READ) {
- tn.addr_read = address;
- if (wp_flags & BP_MEM_READ) {
- tn.addr_read |= TLB_WATCHPOINT;
- }
- } else {
- tn.addr_read = -1;
- }
- if (prot & PAGE_EXEC) {
- tn.addr_code = address;
- } else {
- tn.addr_code = -1;
- }
+ tlb_set_compare(full, &tn, addr_page, read_flags,
+ MMU_INST_FETCH, prot & PAGE_EXEC);
- tn.addr_write = -1;
- if (prot & PAGE_WRITE) {
- tn.addr_write = write_address;
- if (prot & PAGE_WRITE_INV) {
- tn.addr_write |= TLB_INVALID_MASK;
- }
- if (wp_flags & BP_MEM_WRITE) {
- tn.addr_write |= TLB_WATCHPOINT;
- }
+ if (wp_flags & BP_MEM_READ) {
+ read_flags |= TLB_WATCHPOINT;
}
+ tlb_set_compare(full, &tn, addr_page, read_flags,
+ MMU_DATA_LOAD, prot & PAGE_READ);
+
+ if (prot & PAGE_WRITE_INV) {
+ write_flags |= TLB_INVALID_MASK;
+ }
+ if (wp_flags & BP_MEM_WRITE) {
+ write_flags |= TLB_WATCHPOINT;
+ }
+ tlb_set_compare(full, &tn, addr_page, write_flags,
+ MMU_DATA_STORE, prot & PAGE_WRITE);
copy_tlb_helper_locked(te, &tn);
tlb_n_used_entries_inc(env, mmu_idx);
@@ -1509,7 +1519,8 @@ static int probe_access_internal(CPUArchState *env, vaddr addr,
CPUTLBEntry *entry = tlb_entry(env, mmu_idx, addr);
uint64_t tlb_addr = tlb_read_idx(entry, access_type);
vaddr page_addr = addr & TARGET_PAGE_MASK;
- int flags = TLB_FLAGS_MASK;
+ int flags = TLB_FLAGS_MASK & ~TLB_FORCE_SLOW;
+ CPUTLBEntryFull *full;
if (!tlb_hit_page(tlb_addr, page_addr)) {
if (!victim_tlb_hit(env, mmu_idx, index, access_type, page_addr)) {
@@ -1538,7 +1549,8 @@ static int probe_access_internal(CPUArchState *env, vaddr addr,
}
flags &= tlb_addr;
- *pfull = &env_tlb(env)->d[mmu_idx].fulltlb[index];
+ *pfull = full = &env_tlb(env)->d[mmu_idx].fulltlb[index];
+ flags |= full->slow_flags[access_type];
/* Fold all "mmio-like" bits into TLB_MMIO. This is not RAM. */
if (unlikely(flags & ~(TLB_WATCHPOINT | TLB_NOTDIRTY))) {
@@ -1761,6 +1773,8 @@ static bool mmu_lookup1(CPUArchState *env, MMULookupPageData *data,
CPUTLBEntry *entry = tlb_entry(env, mmu_idx, addr);
uint64_t tlb_addr = tlb_read_idx(entry, access_type);
bool maybe_resized = false;
+ CPUTLBEntryFull *full;
+ int flags;
/* If the TLB entry is for a different page, reload and try again. */
if (!tlb_hit(tlb_addr, addr)) {
@@ -1774,8 +1788,12 @@ static bool mmu_lookup1(CPUArchState *env, MMULookupPageData *data,
tlb_addr = tlb_read_idx(entry, access_type) & ~TLB_INVALID_MASK;
}
- data->flags = tlb_addr & TLB_FLAGS_MASK;
- data->full = &env_tlb(env)->d[mmu_idx].fulltlb[index];
+ full = &env_tlb(env)->d[mmu_idx].fulltlb[index];
+ flags = tlb_addr & (TLB_FLAGS_MASK & ~TLB_FORCE_SLOW);
+ flags |= full->slow_flags[access_type];
+
+ data->full = full;
+ data->flags = flags;
/* Compute haddr speculatively; depending on flags it might be invalid. */
data->haddr = (void *)((uintptr_t)addr + entry->addend);
--
2.34.1
^ permalink raw reply related [flat|nested] 28+ messages in thread* [PULL 21/22] accel/tcg: Move TLB_WATCHPOINT to TLB_SLOW_FLAGS_MASK
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
` (19 preceding siblings ...)
2023-06-26 15:39 ` [PULL 20/22] accel/tcg: Store some tlb flags in CPUTLBEntryFull Richard Henderson
@ 2023-06-26 15:39 ` Richard Henderson
2023-06-26 15:39 ` [PULL 22/22] accel/tcg: Renumber TLB_DISCARD_WRITE Richard Henderson
2023-06-26 18:11 ` [PULL 00/22] tcg patch queue Richard Henderson
22 siblings, 0 replies; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 15:39 UTC (permalink / raw)
To: qemu-devel; +Cc: Philippe Mathieu-Daudé
This frees up one bit of the primary tlb flags without
impacting the TLB_NOTDIRTY logic.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
include/exec/cpu-all.h | 8 ++++----
accel/tcg/cputlb.c | 18 ++++++++++++++----
2 files changed, 18 insertions(+), 8 deletions(-)
diff --git a/include/exec/cpu-all.h b/include/exec/cpu-all.h
index 4422f4bb07..b5618613cc 100644
--- a/include/exec/cpu-all.h
+++ b/include/exec/cpu-all.h
@@ -325,8 +325,6 @@ CPUArchState *cpu_copy(CPUArchState *env);
#define TLB_NOTDIRTY (1 << (TARGET_PAGE_BITS_MIN - 2))
/* Set if TLB entry is an IO callback. */
#define TLB_MMIO (1 << (TARGET_PAGE_BITS_MIN - 3))
-/* Set if TLB entry contains a watchpoint. */
-#define TLB_WATCHPOINT (1 << (TARGET_PAGE_BITS_MIN - 4))
/* Set if the slow path must be used; more flags in CPUTLBEntryFull. */
#define TLB_FORCE_SLOW (1 << (TARGET_PAGE_BITS_MIN - 5))
/* Set if TLB entry writes ignored. */
@@ -338,7 +336,7 @@ CPUArchState *cpu_copy(CPUArchState *env);
*/
#define TLB_FLAGS_MASK \
(TLB_INVALID_MASK | TLB_NOTDIRTY | TLB_MMIO \
- | TLB_WATCHPOINT | TLB_FORCE_SLOW | TLB_DISCARD_WRITE)
+ | TLB_FORCE_SLOW | TLB_DISCARD_WRITE)
/*
* Flags stored in CPUTLBEntryFull.slow_flags[x].
@@ -346,8 +344,10 @@ CPUArchState *cpu_copy(CPUArchState *env);
*/
/* Set if TLB entry requires byte swap. */
#define TLB_BSWAP (1 << 0)
+/* Set if TLB entry contains a watchpoint. */
+#define TLB_WATCHPOINT (1 << 1)
-#define TLB_SLOW_FLAGS_MASK TLB_BSWAP
+#define TLB_SLOW_FLAGS_MASK (TLB_BSWAP | TLB_WATCHPOINT)
/* The two sets of flags must not overlap. */
QEMU_BUILD_BUG_ON(TLB_FLAGS_MASK & TLB_SLOW_FLAGS_MASK);
diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c
index 3671846744..5b51eff5a4 100644
--- a/accel/tcg/cputlb.c
+++ b/accel/tcg/cputlb.c
@@ -1981,7 +1981,7 @@ static void *atomic_mmu_lookup(CPUArchState *env, vaddr addr, MemOpIdx oi,
*/
goto stop_the_world;
}
- /* Collect TLB_WATCHPOINT for read. */
+ /* Collect tlb flags for read. */
tlb_addr |= tlbe->addr_read;
/* Notice an IO access or a needs-MMU-lookup access */
@@ -1998,9 +1998,19 @@ static void *atomic_mmu_lookup(CPUArchState *env, vaddr addr, MemOpIdx oi,
notdirty_write(env_cpu(env), addr, size, full, retaddr);
}
- if (unlikely(tlb_addr & TLB_WATCHPOINT)) {
- cpu_check_watchpoint(env_cpu(env), addr, size, full->attrs,
- BP_MEM_READ | BP_MEM_WRITE, retaddr);
+ if (unlikely(tlb_addr & TLB_FORCE_SLOW)) {
+ int wp_flags = 0;
+
+ if (full->slow_flags[MMU_DATA_STORE] & TLB_WATCHPOINT) {
+ wp_flags |= BP_MEM_WRITE;
+ }
+ if (full->slow_flags[MMU_DATA_LOAD] & TLB_WATCHPOINT) {
+ wp_flags |= BP_MEM_READ;
+ }
+ if (wp_flags) {
+ cpu_check_watchpoint(env_cpu(env), addr, size,
+ full->attrs, wp_flags, retaddr);
+ }
}
return hostaddr;
--
2.34.1
^ permalink raw reply related [flat|nested] 28+ messages in thread* [PULL 22/22] accel/tcg: Renumber TLB_DISCARD_WRITE
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
` (20 preceding siblings ...)
2023-06-26 15:39 ` [PULL 21/22] accel/tcg: Move TLB_WATCHPOINT to TLB_SLOW_FLAGS_MASK Richard Henderson
@ 2023-06-26 15:39 ` Richard Henderson
2023-06-26 18:11 ` [PULL 00/22] tcg patch queue Richard Henderson
22 siblings, 0 replies; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 15:39 UTC (permalink / raw)
To: qemu-devel; +Cc: Philippe Mathieu-Daudé
Move to fill a hole in the set of bits.
Reduce the total number of tlb bits by 1.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
include/exec/cpu-all.h | 4 ++--
tcg/tcg-op-ldst.c | 2 +-
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/include/exec/cpu-all.h b/include/exec/cpu-all.h
index b5618613cc..8018ce783e 100644
--- a/include/exec/cpu-all.h
+++ b/include/exec/cpu-all.h
@@ -325,10 +325,10 @@ CPUArchState *cpu_copy(CPUArchState *env);
#define TLB_NOTDIRTY (1 << (TARGET_PAGE_BITS_MIN - 2))
/* Set if TLB entry is an IO callback. */
#define TLB_MMIO (1 << (TARGET_PAGE_BITS_MIN - 3))
+/* Set if TLB entry writes ignored. */
+#define TLB_DISCARD_WRITE (1 << (TARGET_PAGE_BITS_MIN - 4))
/* Set if the slow path must be used; more flags in CPUTLBEntryFull. */
#define TLB_FORCE_SLOW (1 << (TARGET_PAGE_BITS_MIN - 5))
-/* Set if TLB entry writes ignored. */
-#define TLB_DISCARD_WRITE (1 << (TARGET_PAGE_BITS_MIN - 6))
/*
* Use this mask to check interception with an alignment mask
diff --git a/tcg/tcg-op-ldst.c b/tcg/tcg-op-ldst.c
index a4f51bfb6e..0fcc1618e5 100644
--- a/tcg/tcg-op-ldst.c
+++ b/tcg/tcg-op-ldst.c
@@ -39,7 +39,7 @@ static void check_max_alignment(unsigned a_bits)
* The requested alignment cannot overlap the TLB flags.
* FIXME: Must keep the count up-to-date with "exec/cpu-all.h".
*/
- tcg_debug_assert(a_bits + 6 <= tcg_ctx->page_bits);
+ tcg_debug_assert(a_bits + 5 <= tcg_ctx->page_bits);
#endif
}
--
2.34.1
^ permalink raw reply related [flat|nested] 28+ messages in thread* Re: [PULL 00/22] tcg patch queue
2023-06-26 15:39 [PULL 00/22] tcg patch queue Richard Henderson
` (21 preceding siblings ...)
2023-06-26 15:39 ` [PULL 22/22] accel/tcg: Renumber TLB_DISCARD_WRITE Richard Henderson
@ 2023-06-26 18:11 ` Richard Henderson
22 siblings, 0 replies; 28+ messages in thread
From: Richard Henderson @ 2023-06-26 18:11 UTC (permalink / raw)
To: qemu-devel
On 6/26/23 17:39, Richard Henderson wrote:
> The following changes since commit 390e8fc6b0e7b521c9eceb8dfe0958e141009ab9:
>
> Merge tag 'for_upstream' ofhttps://git.kernel.org/pub/scm/virt/kvm/mst/qemu into staging (2023-06-26 16:05:45 +0200)
>
> are available in the Git repository at:
>
> https://gitlab.com/rth7680/qemu.git tags/pull-tcg-20230626
>
> for you to fetch changes up to a0eaae08c7c6a59c185cf646b02f4167b2ac6ec0:
>
> accel/tcg: Renumber TLB_DISCARD_WRITE (2023-06-26 17:33:00 +0200)
>
> ----------------------------------------------------------------
> accel/tcg: Replace target_ulong in some APIs
> accel/tcg: Remove CONFIG_PROFILER
> accel/tcg: Store some tlb flags in CPUTLBEntryFull
> tcg: Issue memory barriers as required for the guest memory model
> tcg: Fix temporary variable in tcg_gen_gvec_andcs
Applied, thanks. Please update https://wiki.qemu.org/ChangeLog/8.1 as appropriate.
r~
^ permalink raw reply [flat|nested] 28+ messages in thread