[Qemu-devel] [PATCH 17/36] cputlb: Byte swap memory transaction attribute

Richard Henderson posted 36 patches 5 years, 7 months ago
Maintainers: Eduardo Habkost <ehabkost@redhat.com>, Claudio Fontana <claudio.fontana@huawei.com>, Stefan Hajnoczi <stefanha@redhat.com>, Alex Williamson <alex.williamson@redhat.com>, Palmer Dabbelt <palmer@sifive.com>, Richard Henderson <rth@twiddle.net>, Laurent Vivier <laurent@vivier.eu>, Halil Pasic <pasic@linux.ibm.com>, Alistair Francis <Alistair.Francis@wdc.com>, Peter Maydell <peter.maydell@linaro.org>, Paolo Bonzini <pbonzini@redhat.com>, "Edgar E. Iglesias" <edgar.iglesias@gmail.com>, Bastian Koppelmann <kbastian@mail.uni-paderborn.de>, "Michael S. Tsirkin" <mst@redhat.com>, Riku Voipio <riku.voipio@iki.fi>, Cornelia Huck <cohuck@redhat.com>, David Gibson <david@gibson.dropbear.id.au>, Aleksandar Rikalo <arikalo@wavecomp.com>, Aleksandar Markovic <amarkovic@wavecomp.com>, Aurelien Jarno <aurelien@aurel32.net>, Stafford Horne <shorne@gmail.com>, Artyom Tarasenko <atar4qemu@gmail.com>, Andrzej Zaborowski <balrogg@gmail.com>, Marcel Apfelbaum <marcel.apfelbaum@gmail.com>, Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>, Sagar Karandikar <sagark@eecs.berkeley.edu>, Collin Walling <walling@linux.ibm.com>, Christian Borntraeger <borntraeger@de.ibm.com>, David Hildenbrand <david@redhat.com>
There is a newer version of this series
[Qemu-devel] [PATCH 17/36] cputlb: Byte swap memory transaction attribute
Posted by Richard Henderson 5 years, 7 months ago
From: Tony Nguyen <tony.nguyen@bt.com>

Notice new attribute, byte swap, and force the transaction through the
memory slow path.

Required by architectures that can invert endianness of memory
transaction, e.g. SPARC64 has the Invert Endian TTE bit.

Suggested-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Tony Nguyen <tony.nguyen@bt.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <2a10a1f1c00a894af1212c8f68ef09c2966023c1.1566466906.git.tony.nguyen@bt.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
 include/exec/memattrs.h |  2 ++
 accel/tcg/cputlb.c      | 12 ++++++++++++
 2 files changed, 14 insertions(+)

diff --git a/include/exec/memattrs.h b/include/exec/memattrs.h
index d4a3477d71..95f2d20d55 100644
--- a/include/exec/memattrs.h
+++ b/include/exec/memattrs.h
@@ -37,6 +37,8 @@ typedef struct MemTxAttrs {
     unsigned int user:1;
     /* Requester ID (for MSI for example) */
     unsigned int requester_id:16;
+    /* Invert endianness for this page */
+    unsigned int byte_swap:1;
     /*
      * The following are target-specific page-table bits.  These are not
      * related to actual memory transactions at all.  However, this structure
diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c
index 3c9e634d99..d9787cc893 100644
--- a/accel/tcg/cputlb.c
+++ b/accel/tcg/cputlb.c
@@ -738,6 +738,10 @@ void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr,
          */
         address |= TLB_RECHECK;
     }
+    if (attrs.byte_swap) {
+        /* Force the access through the I/O slow path.  */
+        address |= TLB_MMIO;
+    }
     if (!memory_region_is_ram(section->mr) &&
         !memory_region_is_romd(section->mr)) {
         /* IO memory case */
@@ -891,6 +895,10 @@ static uint64_t io_readx(CPUArchState *env, CPUIOTLBEntry *iotlbentry,
     bool locked = false;
     MemTxResult r;
 
+    if (iotlbentry->attrs.byte_swap) {
+        op ^= MO_BSWAP;
+    }
+
     section = iotlb_to_section(cpu, iotlbentry->addr, iotlbentry->attrs);
     mr = section->mr;
     mr_offset = (iotlbentry->addr & TARGET_PAGE_MASK) + addr;
@@ -933,6 +941,10 @@ static void io_writex(CPUArchState *env, CPUIOTLBEntry *iotlbentry,
     bool locked = false;
     MemTxResult r;
 
+    if (iotlbentry->attrs.byte_swap) {
+        op ^= MO_BSWAP;
+    }
+
     section = iotlb_to_section(cpu, iotlbentry->addr, iotlbentry->attrs);
     mr = section->mr;
     mr_offset = (iotlbentry->addr & TARGET_PAGE_MASK) + addr;
-- 
2.17.1