* [PATCH] parisc: Don't enforce DMA completion order in cache flushes
@ 2022-03-30 14:42 John David Anglin
0 siblings, 0 replies; only message in thread
From: John David Anglin @ 2022-03-30 14:42 UTC (permalink / raw)
To: linux-parisc; +Cc: Helge Deller, Deller, James Bottomley
[-- Attachment #1: Type: text/plain, Size: 2244 bytes --]
The only place we need to ensure all outstanding cache coherence
operations are complete is in invalidate_kernel_vmap_range. All
parisc drivers synchronize DMA operations internally and do not
call invalidate_kernel_vmap_range. We only need this for non-coherent
I/O operations.
Signed-off-by: John David Anglin <dave.anglin@bell.net>
---
diff --git a/arch/parisc/include/asm/cache.h b/arch/parisc/include/asm/cache.h
index 5032e758594e..e23d06b51a20 100644
--- a/arch/parisc/include/asm/cache.h
+++ b/arch/parisc/include/asm/cache.h
@@ -54,6 +54,7 @@ void parisc_setup_cache_timing(void);
#define asm_io_sync() asm volatile("sync" \
ALTERNATIVE(ALT_COND_NO_DCACHE, INSN_NOP) \
ALTERNATIVE(ALT_COND_NO_IOC_FDC, INSN_NOP) :::"memory")
+#define asm_syncdma() asm volatile("syncdma" :::"memory")
#endif /* ! __ASSEMBLY__ */
diff --git a/arch/parisc/kernel/cache.c b/arch/parisc/kernel/cache.c
index 533c5643c3fe..7eb3467a0432 100644
--- a/arch/parisc/kernel/cache.c
+++ b/arch/parisc/kernel/cache.c
@@ -642,6 +642,9 @@ void invalidate_kernel_vmap_range(void *vaddr, int size)
unsigned long start = (unsigned long)vaddr;
unsigned long end = start + size;
+ /* Ensure DMA is complete */
+ asm_syncdma();
+
if ((!IS_ENABLED(CONFIG_SMP) || !arch_irqs_disabled()) &&
(unsigned long)size >= parisc_cache_flush_threshold) {
flush_tlb_kernel_range(start, end);
diff --git a/arch/parisc/kernel/pacache.S b/arch/parisc/kernel/pacache.S
index b4c3f01e2399..8569141e3e67 100644
--- a/arch/parisc/kernel/pacache.S
+++ b/arch/parisc/kernel/pacache.S
@@ -300,7 +300,6 @@ fdoneloop2:
fdce,m %arg1(%sr1, %arg0) /* Fdce for one loop */
fdsync:
- syncdma
sync
mtsm %r22 /* restore I-bit */
89: ALTERNATIVE(88b, 89b, ALT_COND_NO_DCACHE, INSN_NOP)
@@ -1098,7 +1097,6 @@ ENTRY_CFI(flush_kernel_dcache_range_asm)
sync
89: ALTERNATIVE(88b, 89b, ALT_COND_NO_DCACHE, INSN_NOP)
- syncdma
bv %r0(%r2)
nop
ENDPROC_CFI(flush_kernel_dcache_range_asm)
@@ -1140,7 +1138,6 @@ ENTRY_CFI(purge_kernel_dcache_range_asm)
sync
89: ALTERNATIVE(88b, 89b, ALT_COND_NO_DCACHE, INSN_NOP)
- syncdma
bv %r0(%r2)
nop
ENDPROC_CFI(purge_kernel_dcache_range_asm)
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
^ permalink raw reply related [flat|nested] only message in thread
only message in thread, other threads:[~2022-03-30 14:50 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2022-03-30 14:42 [PATCH] parisc: Don't enforce DMA completion order in cache flushes John David Anglin
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox