Skip to contentSkip to conversation

x86emul: pad blob-execution "okay" messages

PATCH In review Jan Beulich 1 participant 3 messages Jun 2, 2021 – Jul 6, 2021

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 x86emul: pad blob-execution "okay" messages

We already do so in the native execution case, and a few descriptions (I did notice this with SHA ones) are short enough for the output to look slightly odd.

---
Many descriptions are longer than 37 characters, so I wonder whether we wouldn't want to bump the padding to 50, 60, or even 70. And this perhaps despite then going out of sync with the individual insn tests earlier on (which I wouldn't want to touch).

tools/tests/x86_emulator/test_x86_emulator.c +18 -6
@@ -5181,6 +5181,8 @@ int main(int argc, char **argv)
 
     for ( j = 0; j < ARRAY_SIZE(blobs); j++ )
     {
+        unsigned int nr;
+
         if ( blobs[j].check_cpu && !blobs[j].check_cpu() )
             continue;
 
@@ -5196,7 +5198,8 @@ int main(int argc, char **argv)
 
         if ( ctxt.addr_size == sizeof(void *) * CHAR_BIT )
         {
-            i = printf("Testing %s native execution...", blobs[j].name);
+            nr = printf("Testing %s native execution...", blobs[j].name);
+
             if ( blobs[j].set_regs )
                 blobs[j].set_regs(&regs);
             asm volatile (
@@ -5212,11 +5215,13 @@ int main(int argc, char **argv)
             );
             if ( !blobs[j].check_regs(&regs) )
                 goto fail;
-            printf("%*sokay\n", i < 40 ? 40 - i : 0, "");
+
+            printf("%*sokay\n", nr < 40 ? 40 - nr : 0, "");
         }
 
-        printf("Testing %s %u-bit code sequence",
-               blobs[j].name, ctxt.addr_size);
+        nr = printf("Testing %s %u-bit code sequence",
+                    blobs[j].name, ctxt.addr_size);
+
         if ( blobs[j].set_regs )
             blobs[j].set_regs(&regs);
         regs.eip = (unsigned long)res;
@@ -5233,7 +5238,10 @@ int main(int argc, char **argv)
                 regs.eip < (unsigned long)res + blobs[j].size )
         {
             if ( (i++ & 8191) == 0 )
+            {
                 printf(".");
+                ++nr;
+            }
             rc = x86_emulate(&ctxt, &emulops);
             if ( rc != X86EMUL_OKAY )
             {
@@ -5242,13 +5250,17 @@ int main(int argc, char **argv)
                 return 1;
             }
         }
-        for ( ; i < 2 * 8192; i += 8192 )
+        for ( ; i < 2 * 8192; i += 8192 ) {
             printf(".");
+            ++nr;
+        }
+
         if ( (regs.eip != 0x12345678) ||
              (regs.esp != ((unsigned long)res + MMAP_SZ)) ||
              !blobs[j].check_regs(&regs) )
             goto fail;
-        printf("okay\n");
+
+        printf("%*sokay\n", nr < 40 ? 40 - nr : 0, "");
     }
 
     return 0;
#1 lore raw

Ping: [PATCH] x86emul: pad blob-execution "okay" messages

Quoted 5 lines · Jan Beulich

We already do so in the native execution case, and a few descriptions (I did notice this with SHA ones) are short enough for the output to look slightly odd.

Again - anyone?

Thanks, Jan

Quoted 77 lines

---
Many descriptions are longer than 37 characters, so I wonder whether we wouldn't want to bump the padding to 50, 60, or even 70. And this perhaps despite then going out of sync with the individual insn tests earlier on (which I wouldn't want to touch).

tools/tests/x86_emulator/test_x86_emulator.c +18 -6
@@ -5181,6 +5181,8 @@ int main(int argc, char **argv)
 
     for ( j = 0; j < ARRAY_SIZE(blobs); j++ )
     {
+        unsigned int nr;
+
         if ( blobs[j].check_cpu && !blobs[j].check_cpu() )
             continue;
 
@@ -5196,7 +5198,8 @@ int main(int argc, char **argv)
 
         if ( ctxt.addr_size == sizeof(void *) * CHAR_BIT )
         {
-            i = printf("Testing %s native execution...", blobs[j].name);
+            nr = printf("Testing %s native execution...", blobs[j].name);
+
             if ( blobs[j].set_regs )
                 blobs[j].set_regs(&regs);
             asm volatile (
@@ -5212,11 +5215,13 @@ int main(int argc, char **argv)
             );
             if ( !blobs[j].check_regs(&regs) )
                 goto fail;
-            printf("%*sokay\n", i < 40 ? 40 - i : 0, "");
+
+            printf("%*sokay\n", nr < 40 ? 40 - nr : 0, "");
         }
 
-        printf("Testing %s %u-bit code sequence",
-               blobs[j].name, ctxt.addr_size);
+        nr = printf("Testing %s %u-bit code sequence",
+                    blobs[j].name, ctxt.addr_size);
+
         if ( blobs[j].set_regs )
             blobs[j].set_regs(&regs);
         regs.eip = (unsigned long)res;
@@ -5233,7 +5238,10 @@ int main(int argc, char **argv)
                 regs.eip < (unsigned long)res + blobs[j].size )
         {
             if ( (i++ & 8191) == 0 )
+            {
                 printf(".");
+                ++nr;
+            }
             rc = x86_emulate(&ctxt, &emulops);
             if ( rc != X86EMUL_OKAY )
             {
@@ -5242,13 +5250,17 @@ int main(int argc, char **argv)
                 return 1;
             }
         }
-        for ( ; i < 2 * 8192; i += 8192 )
+        for ( ; i < 2 * 8192; i += 8192 ) {
             printf(".");
+            ++nr;
+        }
+
         if ( (regs.eip != 0x12345678) ||
              (regs.esp != ((unsigned long)res + MMAP_SZ)) ||
              !blobs[j].check_regs(&regs) )
             goto fail;
-        printf("okay\n");
+
+        printf("%*sokay\n", nr < 40 ? 40 - nr : 0, "");
     }
 
     return 0;
#2 lore raw

Re: Ping: [PATCH] x86emul: pad blob-execution "okay" messages

Quoted 8 lines · Jan Beulich
Quoted 5 lines · Jan Beulich

We already do so in the native execution case, and a few descriptions (I did notice this with SHA ones) are short enough for the output to look slightly odd.

Again - anyone?

Okay, this is trivial enough to fall under "lazy consensus" if I don't hear back by Thursday (i.e. including the Community Call as an option to voice objections).

Jan