All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH 0/2] target/cris: Pass MMUAccessType to cris_mmu_translate()
@ 2021-01-28  0:32 Philippe Mathieu-Daudé
  2021-01-28  0:32 ` [PATCH 1/2] target/cris: Use MMUAccessType enum type when possible Philippe Mathieu-Daudé
  2021-01-28  0:32 ` [PATCH 2/2] target/cris: Let cris_mmu_translate() use MMUAccessType access_type Philippe Mathieu-Daudé
  0 siblings, 2 replies; 7+ messages in thread
From: Philippe Mathieu-Daudé @ 2021-01-28  0:32 UTC (permalink / raw)
  To: qemu-devel
  Cc: Edgar E. Iglesias, Richard Henderson, Philippe Mathieu-Daudé,
	Joe Komlodi

Taking notes while reviewing commit 671a0a1265a
("use MMUAccessType instead of int in mmu_translate").

Philippe Mathieu-Daudé (2):
  target/cris: Use MMUAccessType enum type when possible
  target/cris: Let cris_mmu_translate() use MMUAccessType access_type

 target/cris/mmu.h    |  2 +-
 target/cris/helper.c |  4 ++--
 target/cris/mmu.c    | 31 +++++++++++++++----------------
 3 files changed, 18 insertions(+), 19 deletions(-)

-- 
2.26.2



^ permalink raw reply	[flat|nested] 7+ messages in thread

* [PATCH 1/2] target/cris: Use MMUAccessType enum type when possible
  2021-01-28  0:32 [PATCH 0/2] target/cris: Pass MMUAccessType to cris_mmu_translate() Philippe Mathieu-Daudé
@ 2021-01-28  0:32 ` Philippe Mathieu-Daudé
  2021-02-04  2:45   ` Richard Henderson
  2021-01-28  0:32 ` [PATCH 2/2] target/cris: Let cris_mmu_translate() use MMUAccessType access_type Philippe Mathieu-Daudé
  1 sibling, 1 reply; 7+ messages in thread
From: Philippe Mathieu-Daudé @ 2021-01-28  0:32 UTC (permalink / raw)
  To: qemu-devel
  Cc: Edgar E. Iglesias, Richard Henderson, Philippe Mathieu-Daudé,
	Joe Komlodi

Replace the 0/1/2 magic values by the corresponding MMUAccessType.
We can remove a comment as enum names are self explicit.

Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
---
 target/cris/helper.c |  4 ++--
 target/cris/mmu.c    | 13 ++++++-------
 2 files changed, 8 insertions(+), 9 deletions(-)

diff --git a/target/cris/helper.c b/target/cris/helper.c
index ed45c3d9b70..cc110f1154d 100644
--- a/target/cris/helper.c
+++ b/target/cris/helper.c
@@ -274,10 +274,10 @@ hwaddr cris_cpu_get_phys_page_debug(CPUState *cs, vaddr addr)
     struct cris_mmu_result res;
     int miss;
 
-    miss = cris_mmu_translate(&res, &cpu->env, addr, 0, 0, 1);
+    miss = cris_mmu_translate(&res, &cpu->env, addr, MMU_DATA_LOAD, 0, 1);
     /* If D TLB misses, try I TLB.  */
     if (miss) {
-        miss = cris_mmu_translate(&res, &cpu->env, addr, 2, 0, 1);
+        miss = cris_mmu_translate(&res, &cpu->env, addr, MMU_INST_FETCH, 0, 1);
     }
 
     if (!miss) {
diff --git a/target/cris/mmu.c b/target/cris/mmu.c
index a279b7f1b60..294de7dffd5 100644
--- a/target/cris/mmu.c
+++ b/target/cris/mmu.c
@@ -152,15 +152,15 @@ static int cris_mmu_translate_page(struct cris_mmu_result *res,
     pid = env->pregs[PR_PID] & 0xff;
 
     switch (rw) {
-    case 2:
+    case MMU_INST_FETCH:
         rwcause = CRIS_MMU_ERR_EXEC;
         mmu = 0;
         break;
-    case 1:
+    case MMU_DATA_STORE:
         rwcause = CRIS_MMU_ERR_WRITE;
         break;
     default:
-    case 0:
+    case MMU_DATA_LOAD:
         rwcause = CRIS_MMU_ERR_READ;
         break;
     }
@@ -219,13 +219,13 @@ static int cris_mmu_translate_page(struct cris_mmu_result *res,
                      vaddr, lo, env->pc));
             match = 0;
             res->bf_vec = vect_base + 2;
-        } else if (rw == 1 && cfg_w && !tlb_w) {
+        } else if (rw == MMU_DATA_STORE && cfg_w && !tlb_w) {
             D(printf("tlb: write protected %x lo=%x pc=%x\n",
                      vaddr, lo, env->pc));
             match = 0;
             /* write accesses never go through the I mmu.  */
             res->bf_vec = vect_base + 3;
-        } else if (rw == 2 && cfg_x && !tlb_x) {
+        } else if (rw == MMU_INST_FETCH && cfg_x && !tlb_x) {
             D(printf("tlb: exec protected %x lo=%x pc=%x\n",
                      vaddr, lo, env->pc));
             match = 0;
@@ -329,8 +329,7 @@ int cris_mmu_translate(struct cris_mmu_result *res,
 
     old_srs = env->pregs[PR_SRS];
 
-    /* rw == 2 means exec, map the access to the insn mmu.  */
-    env->pregs[PR_SRS] = rw == 2 ? 1 : 2;
+    env->pregs[PR_SRS] = rw == MMU_INST_FETCH ? 1 : 2;
 
     if (!cris_mmu_enabled(env->sregs[SFR_RW_GC_CFG])) {
         res->phy = vaddr;
-- 
2.26.2



^ permalink raw reply related	[flat|nested] 7+ messages in thread

* [PATCH 2/2] target/cris: Let cris_mmu_translate() use MMUAccessType access_type
  2021-01-28  0:32 [PATCH 0/2] target/cris: Pass MMUAccessType to cris_mmu_translate() Philippe Mathieu-Daudé
  2021-01-28  0:32 ` [PATCH 1/2] target/cris: Use MMUAccessType enum type when possible Philippe Mathieu-Daudé
@ 2021-01-28  0:32 ` Philippe Mathieu-Daudé
  2021-02-04  2:46   ` Richard Henderson
  1 sibling, 1 reply; 7+ messages in thread
From: Philippe Mathieu-Daudé @ 2021-01-28  0:32 UTC (permalink / raw)
  To: qemu-devel
  Cc: Edgar E. Iglesias, Richard Henderson, Philippe Mathieu-Daudé,
	Joe Komlodi

All callers of cris_mmu_translate() provide a MMUAccessType
type. Let the prototype use it as argument, as it is stricter
than an integer. We can remove the documentation as enum
names are self explicit.

Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
---
 target/cris/mmu.h |  2 +-
 target/cris/mmu.c | 24 ++++++++++++------------
 2 files changed, 13 insertions(+), 13 deletions(-)

diff --git a/target/cris/mmu.h b/target/cris/mmu.h
index 9ab1642b964..d57386ec6cd 100644
--- a/target/cris/mmu.h
+++ b/target/cris/mmu.h
@@ -17,6 +17,6 @@ void cris_mmu_init(CPUCRISState *env);
 void cris_mmu_flush_pid(CPUCRISState *env, uint32_t pid);
 int cris_mmu_translate(struct cris_mmu_result *res,
                        CPUCRISState *env, uint32_t vaddr,
-                       int rw, int mmu_idx, int debug);
+                       MMUAccessType access_type, int mmu_idx, int debug);
 
 #endif
diff --git a/target/cris/mmu.c b/target/cris/mmu.c
index 294de7dffd5..b574ec6e5b9 100644
--- a/target/cris/mmu.c
+++ b/target/cris/mmu.c
@@ -129,10 +129,10 @@ static void dump_tlb(CPUCRISState *env, int mmu)
 }
 #endif
 
-/* rw 0 = read, 1 = write, 2 = exec.  */
 static int cris_mmu_translate_page(struct cris_mmu_result *res,
-				   CPUCRISState *env, uint32_t vaddr,
-				   int rw, int usermode, int debug)
+                                   CPUCRISState *env, uint32_t vaddr,
+                                   MMUAccessType access_type,
+                                   int usermode, int debug)
 {
     unsigned int vpage;
     unsigned int idx;
@@ -151,7 +151,7 @@ static int cris_mmu_translate_page(struct cris_mmu_result *res,
     r_cfg = env->sregs[SFR_RW_MM_CFG];
     pid = env->pregs[PR_PID] & 0xff;
 
-    switch (rw) {
+    switch (access_type) {
     case MMU_INST_FETCH:
         rwcause = CRIS_MMU_ERR_EXEC;
         mmu = 0;
@@ -219,13 +219,13 @@ static int cris_mmu_translate_page(struct cris_mmu_result *res,
                      vaddr, lo, env->pc));
             match = 0;
             res->bf_vec = vect_base + 2;
-        } else if (rw == MMU_DATA_STORE && cfg_w && !tlb_w) {
+        } else if (access_type == MMU_DATA_STORE && cfg_w && !tlb_w) {
             D(printf("tlb: write protected %x lo=%x pc=%x\n",
                      vaddr, lo, env->pc));
             match = 0;
             /* write accesses never go through the I mmu.  */
             res->bf_vec = vect_base + 3;
-        } else if (rw == MMU_INST_FETCH && cfg_x && !tlb_x) {
+        } else if (access_type == MMU_INST_FETCH && cfg_x && !tlb_x) {
             D(printf("tlb: exec protected %x lo=%x pc=%x\n",
                      vaddr, lo, env->pc));
             match = 0;
@@ -272,9 +272,9 @@ static int cris_mmu_translate_page(struct cris_mmu_result *res,
         D(printf("refill vaddr=%x pc=%x\n", vaddr, env->pc));
     }
 
-    D(printf("%s rw=%d mtch=%d pc=%x va=%x vpn=%x tlbvpn=%x pfn=%x pid=%x"
+    D(printf("%s access=%u mtch=%d pc=%x va=%x vpn=%x tlbvpn=%x pfn=%x pid=%x"
              " %x cause=%x sel=%x sp=%x %x %x\n",
-             __func__, rw, match, env->pc,
+             __func__, access_type, match, env->pc,
              vaddr, vpage,
              tlb_vpn, tlb_pfn, tlb_pid,
              pid,
@@ -319,8 +319,8 @@ void cris_mmu_flush_pid(CPUCRISState *env, uint32_t pid)
 }
 
 int cris_mmu_translate(struct cris_mmu_result *res,
-		       CPUCRISState *env, uint32_t vaddr,
-		       int rw, int mmu_idx, int debug)
+                       CPUCRISState *env, uint32_t vaddr,
+                       MMUAccessType access_type, int mmu_idx, int debug)
 {
     int seg;
     int miss = 0;
@@ -329,7 +329,7 @@ int cris_mmu_translate(struct cris_mmu_result *res,
 
     old_srs = env->pregs[PR_SRS];
 
-    env->pregs[PR_SRS] = rw == MMU_INST_FETCH ? 1 : 2;
+    env->pregs[PR_SRS] = access_type == MMU_INST_FETCH ? 1 : 2;
 
     if (!cris_mmu_enabled(env->sregs[SFR_RW_GC_CFG])) {
         res->phy = vaddr;
@@ -346,7 +346,7 @@ int cris_mmu_translate(struct cris_mmu_result *res,
         res->phy = base | (0x0fffffff & vaddr);
         res->prot = PAGE_BITS;
     } else {
-        miss = cris_mmu_translate_page(res, env, vaddr, rw,
+        miss = cris_mmu_translate_page(res, env, vaddr, access_type,
                                        is_user, debug);
     }
  done:
-- 
2.26.2



^ permalink raw reply related	[flat|nested] 7+ messages in thread

* Re: [PATCH 1/2] target/cris: Use MMUAccessType enum type when possible
  2021-01-28  0:32 ` [PATCH 1/2] target/cris: Use MMUAccessType enum type when possible Philippe Mathieu-Daudé
@ 2021-02-04  2:45   ` Richard Henderson
  2021-02-04 14:33     ` Edgar E. Iglesias
  0 siblings, 1 reply; 7+ messages in thread
From: Richard Henderson @ 2021-02-04  2:45 UTC (permalink / raw)
  To: Philippe Mathieu-Daudé, qemu-devel; +Cc: Edgar E. Iglesias, Joe Komlodi

On 1/27/21 2:32 PM, Philippe Mathieu-Daudé wrote:
> Replace the 0/1/2 magic values by the corresponding MMUAccessType.
> We can remove a comment as enum names are self explicit.
> 
> Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
> ---
>  target/cris/helper.c |  4 ++--
>  target/cris/mmu.c    | 13 ++++++-------
>  2 files changed, 8 insertions(+), 9 deletions(-)
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>


r~



^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [PATCH 2/2] target/cris: Let cris_mmu_translate() use MMUAccessType access_type
  2021-01-28  0:32 ` [PATCH 2/2] target/cris: Let cris_mmu_translate() use MMUAccessType access_type Philippe Mathieu-Daudé
@ 2021-02-04  2:46   ` Richard Henderson
  2021-02-04 14:33     ` Edgar E. Iglesias
  0 siblings, 1 reply; 7+ messages in thread
From: Richard Henderson @ 2021-02-04  2:46 UTC (permalink / raw)
  To: Philippe Mathieu-Daudé, qemu-devel; +Cc: Edgar E. Iglesias, Joe Komlodi

On 1/27/21 2:32 PM, Philippe Mathieu-Daudé wrote:
> All callers of cris_mmu_translate() provide a MMUAccessType
> type. Let the prototype use it as argument, as it is stricter
> than an integer. We can remove the documentation as enum
> names are self explicit.
> 
> Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
> ---
>  target/cris/mmu.h |  2 +-
>  target/cris/mmu.c | 24 ++++++++++++------------
>  2 files changed, 13 insertions(+), 13 deletions(-)

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>


r~



^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [PATCH 1/2] target/cris: Use MMUAccessType enum type when possible
  2021-02-04  2:45   ` Richard Henderson
@ 2021-02-04 14:33     ` Edgar E. Iglesias
  0 siblings, 0 replies; 7+ messages in thread
From: Edgar E. Iglesias @ 2021-02-04 14:33 UTC (permalink / raw)
  To: Richard Henderson; +Cc: Philippe Mathieu-Daudé, Joe Komlodi, qemu-devel

On Wed, Feb 03, 2021 at 04:45:50PM -1000, Richard Henderson wrote:
> On 1/27/21 2:32 PM, Philippe Mathieu-Daudé wrote:
> > Replace the 0/1/2 magic values by the corresponding MMUAccessType.
> > We can remove a comment as enum names are self explicit.
> > 
> > Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
> > ---
> >  target/cris/helper.c |  4 ++--
> >  target/cris/mmu.c    | 13 ++++++-------
> >  2 files changed, 8 insertions(+), 9 deletions(-)
> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>

Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>

Can this go through the trivial tree?

Thanks,
Edgar


^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [PATCH 2/2] target/cris: Let cris_mmu_translate() use MMUAccessType access_type
  2021-02-04  2:46   ` Richard Henderson
@ 2021-02-04 14:33     ` Edgar E. Iglesias
  0 siblings, 0 replies; 7+ messages in thread
From: Edgar E. Iglesias @ 2021-02-04 14:33 UTC (permalink / raw)
  To: Richard Henderson; +Cc: Philippe Mathieu-Daudé, Joe Komlodi, qemu-devel

On Wed, Feb 03, 2021 at 04:46:28PM -1000, Richard Henderson wrote:
> On 1/27/21 2:32 PM, Philippe Mathieu-Daudé wrote:
> > All callers of cris_mmu_translate() provide a MMUAccessType
> > type. Let the prototype use it as argument, as it is stricter
> > than an integer. We can remove the documentation as enum
> > names are self explicit.
> > 
> > Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
> > ---
> >  target/cris/mmu.h |  2 +-
> >  target/cris/mmu.c | 24 ++++++++++++------------
> >  2 files changed, 13 insertions(+), 13 deletions(-)
> 
> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>

Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>


^ permalink raw reply	[flat|nested] 7+ messages in thread

end of thread, other threads:[~2021-02-04 14:37 UTC | newest]

Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-01-28  0:32 [PATCH 0/2] target/cris: Pass MMUAccessType to cris_mmu_translate() Philippe Mathieu-Daudé
2021-01-28  0:32 ` [PATCH 1/2] target/cris: Use MMUAccessType enum type when possible Philippe Mathieu-Daudé
2021-02-04  2:45   ` Richard Henderson
2021-02-04 14:33     ` Edgar E. Iglesias
2021-01-28  0:32 ` [PATCH 2/2] target/cris: Let cris_mmu_translate() use MMUAccessType access_type Philippe Mathieu-Daudé
2021-02-04  2:46   ` Richard Henderson
2021-02-04 14:33     ` Edgar E. Iglesias

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.