<div>For flush all data cache, the flush action begins from Level 0, then increase cache level by 1.</div><div><br></div><div>But in function v7_flush_dcache_all, the step is 2. IOW, it will just flush 0,2,4,6 level cache.</div>
<div>As the following code, the step is stored in r10, and the increase line is "add r10, r10, #2".</div><div><br></div><div>Is my wrong understanding?</div><div><br></div><div>+ENTRY(v7_flush_dcache_all)</div>
<div>+ mrc p15, 1, r0, c0, c0, 1 @ read clidr</div><div>+ ands r3, r0, #0x7000000 @ extract loc from clidr</div><div>+ mov r3, r3, lsr #23 @ left align loc bit field</div>
<div>+ beq finished @ if loc is 0, then no need to clean</div><div>+ mov r10, #0 @ start clean at cache level 0</div><div>+loop1:</div><div>+ add r2, r10, r10, lsr #1 @ work out 3x current cache level</div>
<div>+ mov r1, r0, lsr r2 @ extract cache type bits from clidr</div><div>+ and r1, r1, #7 @ mask of the bits for current cache only</div><div>+ cmp r1, #2 @ see what cache we have at this level</div>
<div>+ blt skip @ skip if no cache, or just i-cache</div><div>+ mcr p15, 2, r10, c0, c0, 0 @ select current cache level in cssr</div><div>+ isb @ isb to sych the new cssr&csidr</div>
<div>+ mrc p15, 1, r1, c0, c0, 0 @ read the new csidr</div><div>+ and r2, r1, #7 @ extract the length of the cache lines</div><div>+ add r2, r2, #4 @ add 4 (line length offset)</div>
<div>+ ldr r4, =0x3ff</div><div>+ ands r4, r4, r1, lsr #3 @ find maximum number on the way size</div><div>+ clz r5, r4 @ find bit position of way size increment</div>
<div>+ ldr r7, =0x7fff</div><div>+ ands r7, r7, r1, lsr #13 @ extract max number of the index size</div><div>+loop2:</div><div>+ mov r9, r4 @ create working copy of max way size</div>
<div>+loop3:</div><div>+ orr r11, r10, r9, lsl r5 @ factor way and cache number into r11</div><div>+ orr r11, r11, r7, lsl r2 @ factor index number into r11</div><div>+ mcr p15, 0, r11, c7, c14, 2 @ clean & invalidate by set/way</div>
<div>+ subs r9, r9, #1 @ decrement the way</div><div>+ bge loop3</div><div>+ subs r7, r7, #1 @ decrement the index</div><div>+ bge loop2</div>
<div>+skip:</div><div>+ add r10, r10, #2 @ increment cache number</div><div>+ cmp r3, r10</div><div>+ bgt loop1</div><div>+finished:</div><div>+ mov r10, #0 @ swith back to cache level 0</div>
<div>+ mcr p15, 2, r10, c0, c0, 0 @ select current cache level in cssr</div><div>+ isb</div><div>+ mov pc, lr</div><div><br></div>