sravan/system76-coreboot
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

libpayload arm64: Remove the DONT_USE_DC macro

Browse Source 
By default we dont want to use the special DC instruction. Thus getting rid of
the DONT_USE_DC macro and enabling code appropriately in memset.S

BUG=chrome-os-partner:31634
BRANCH=None
TEST=Compiles successfully and memset works fine for mmu init

Change-Id: Icb3193f4f0d122726dcdacbdcacbf53eba30e235
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Original-Commit-Id: 54f639ef2348acab54e32b18f6826a67bf52bc14
Original-Change-Id: Id89ec2c1731d21496eca617a3c03abaf48062908
Original-Signed-off-by: Furquan Shaikh <furquan@google.com>
Original-Reviewed-on: https://chromium-review.googlesource.com/216820
Original-Tested-by: Furquan Shaikh <furquan@chromium.org>
Original-Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Original-Commit-Queue: Furquan Shaikh <furquan@chromium.org>
Reviewed-on: http://review.coreboot.org/8787
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
This commit is contained in:
Furquan Shaikh
2014-09-04 15:05:03 -07:00
committed by Patrick Georgi
parent c10e7f2de9
commit 732f5b5597
1 changed files with 0 additions and 106 deletions
Download Patch File Download Diff File Expand all files Collapse all files

106
payloads/libpayload/arch/arm64/memset.S
View File

@@ -31,16 +31,6 @@
*
*/
/* By default we assume that the DC instruction can be used to zero
data blocks more efficiently. In some circumstances this might be
unsafe, for example in an asymmetric multiprocessor environment with
different DC clear lengths (neither the upper nor lower lengths are
safe to use). The feature can be disabled by defining DONT_USE_DC.
If code may be run in a virtualized environment, then define
MAYBE_VIRT. This will cause the code to cache the system register
values rather than re-reading them each call. */
#define dstin x0
#define val w1
#define count x2
@@ -57,7 +47,6 @@
#define dst x8
#define tmp3w w9
.macro def_fn f p2align=0
.text
.p2align \p2align
@@ -70,9 +59,6 @@ def_fn memset p2align=6
mov dst, dstin /* Preserve return value. */
ands A_lw, val, #255
#ifndef DONT_USE_DC
b.eq .Lzero_mem
#endif
orr A_lw, A_lw, A_lw, lsl #8
orr A_lw, A_lw, A_lw, lsl #16
orr A_l, A_l, A_l, lsl #32
@@ -148,95 +134,3 @@ def_fn memset p2align=6
add dst, dst, #16
b.ne .Ltail63
ret
#ifndef DONT_USE_DC
/* For zeroing memory, check to see if we can use the ZVA feature to
* zero entire 'cache' lines. */
.Lzero_mem:
mov A_l, #0
cmp count, #63
b.le .Ltail_maybe_tiny
neg tmp2, dst
ands tmp2, tmp2, #15
b.eq 1f
sub count, count, tmp2
stp A_l, A_l, [dst]
add dst, dst, tmp2
cmp count, #63
b.le .Ltail63
1:
/* For zeroing small amounts of memory, it's not worth setting up
* the line-clear code. */
cmp count, #128
b.lt .Lnot_short
#ifdef MAYBE_VIRT
/* For efficiency when virtualized, we cache the ZVA capability. */
adrp tmp2, .Lcache_clear
ldr zva_len, [tmp2, #:lo12:.Lcache_clear]
tbnz zva_len, #31, .Lnot_short
cbnz zva_len, .Lzero_by_line
mrs tmp1, dczid_el0
tbz tmp1, #4, 1f
/* ZVA not available. Remember this for next time. */
mov zva_len, #~0
str zva_len, [tmp2, #:lo12:.Lcache_clear]
b .Lnot_short
1:
mov tmp3w, #4
and zva_len, tmp1w, #15 /* Safety: other bits reserved. */
lsl zva_len, tmp3w, zva_len
str zva_len, [tmp2, #:lo12:.Lcache_clear]
#else
mrs tmp1, dczid_el0
tbnz tmp1, #4, .Lnot_short
mov tmp3w, #4
and zva_len, tmp1w, #15 /* Safety: other bits reserved. */
lsl zva_len, tmp3w, zva_len
#endif
.Lzero_by_line:
/* Compute how far we need to go to become suitably aligned. We're
* already at quad-word alignment. */
cmp count, zva_len_x
b.lt .Lnot_short /* Not enough to reach alignment. */
sub zva_bits_x, zva_len_x, #1
neg tmp2, dst
ands tmp2, tmp2, zva_bits_x
b.eq 1f /* Already aligned. */
/* Not aligned, check that there's enough to copy after alignment. */
sub tmp1, count, tmp2
cmp tmp1, #64
ccmp tmp1, zva_len_x, #8, ge /* NZCV=0b1000 */
b.lt .Lnot_short
/* We know that there's at least 64 bytes to zero and that it's safe
* to overrun by 64 bytes. */
mov count, tmp1
2:
stp A_l, A_l, [dst]
stp A_l, A_l, [dst, #16]
stp A_l, A_l, [dst, #32]
subs tmp2, tmp2, #64
stp A_l, A_l, [dst, #48]
add dst, dst, #64
b.ge 2b
/* We've overrun a bit, so adjust dst downwards. */
add dst, dst, tmp2
1:
sub count, count, zva_len_x
3:
dc zva, dst
add dst, dst, zva_len_x
subs count, count, zva_len_x
b.ge 3b
ands count, count, zva_bits_x
b.ne .Ltail_maybe_long
ret
.size memset, .-memset
#ifdef MAYBE_VIRT
.bss
.p2align 2
.Lcache_clear:
.space 4
#endif
#endif /* DONT_USE_DC */