hw/ssi/xilinx_spips: Avoid NULL pointer deference
v4 cover hw/ssi/xilinx_spips: Avoid NULL pointer deference
v1: https://lists.gnu.org/archive/html/qemu-devel/2019-07/msg01238.html
v2: https://lists.gnu.org/archive/html/qemu-devel/2019-07/msg01586.html
v3: https://lists.gnu.org/archive/html/qemu-devel/2019-07/msg01657.html
v4
- address Francisco comments from v3
- RFC avoid out-of-bound
Philippe Mathieu-Daudé (3):
hw/ssi/xilinx_spips: Convert lqspi_read() to read_with_attrs hw/ssi/xilinx_spips: Avoid AXI writes to the LQSPI linear memory hw/ssi/xilinx_spips: Avoid out-of-bound access to lqspi_buf[]
hw/ssi/xilinx_spips.c | 43 +++++++++++++++++++++++++++++++------------ 1 file changed, 31 insertions(+), 12 deletions(-)
Signature · 1 line
2.20.1
PATCH v4 1/3 hw/ssi/xilinx_spips: Convert lqspi_read() to read_with_attrs
In the next commit we will implement the write_with_attrs() handler. To avoid using different APIs, convert the read() handler first.
---
v4: Do not ignore lqspi_read() return value (Francisco)
--- hw/ssi/xilinx_spips.c | 23 +++++++++++------------ 1 file changed, 11 insertions(+), 12 deletions(-)
hw/ssi/xilinx_spips.c +11 -12
@@ -1202,27 +1202,26 @@ static void lqspi_load_cache(void *opaque, hwaddr addr) } } -static uint64_t -lqspi_read (void *opaque,hwaddr addr,unsigned int size) +static MemTxResult lqspi_read (void *opaque,hwaddr addr,uint64_t *value, +unsigned size,MemTxAttrs attrs) { -XilinxQSPIPS *q = opaque; -uint32_t ret; +XilinxQSPIPS *q =XILINX_QSPIPS (opaque);if (addr >= q->lqspi_cached_addr && addr <= q->lqspi_cached_addr +LQSPI_CACHE_SIZE -4 ) {uint8_t *retp = &q->lqspi_buf[addr - q->lqspi_cached_addr]; - ret =cpu_to_le32 (*(uint32_t *)retp); -DB_PRINT_L (1 ,"addr: %08x, data: %08x\n" , (unsigned )addr, - (unsigned )ret); -return ret; - }else { -lqspi_load_cache (opaque, addr); -return lqspi_read (opaque, addr, size); + *value =cpu_to_le32 (*(uint32_t *)retp); +DB_PRINT_L (1 ,"addr: %08" HWADDR_PRIx", data: %08" PRIx64"\n" , + addr, *value); +return MEMTX_OK ; } + +lqspi_load_cache (opaque, addr); +return lqspi_read (opaque, addr, value, size, attrs); }static const MemoryRegionOps lqspi_ops = { - .read = lqspi_read, + .read_with_attrs = lqspi_read, .endianness =DEVICE_NATIVE_ENDIAN , .valid = { .min_access_size =1 ,
Signature · 1 line
2.20.1
Re: [Qemu-devel] [PATCH-for-4.1 v4 1/3] hw/ssi/xilinx_spips: Convert lqspi_read() to read_with_attrs
Quoted 5 lines · Philippe Mathieu-Daudé
In the next commit we will implement the
write_with_attrs()handler. To avoid using different APIs, convert theread()handler first.Signed-off-by Philippe Mathieu-Daudé <[email protected]>
Quoted 52 lines
---
v4: Do not ignorelqspi_read()return value (Francisco)--- hw/ssi/xilinx_spips.c | 23 +++++++++++------------ 1 file changed, 11 insertions(+), 12 deletions(-)hw/ssi/xilinx_spips.c +11
-12@@ -1202,27 +1202,26 @@ static void lqspi_load_cache(void *opaque, hwaddr addr) } } -static uint64_t -lqspi_read (void *opaque,hwaddr addr,unsigned int size) +static MemTxResult lqspi_read (void *opaque,hwaddr addr,uint64_t *value, +unsigned size,MemTxAttrs attrs) { -XilinxQSPIPS *q = opaque; -uint32_t ret; +XilinxQSPIPS *q =XILINX_QSPIPS (opaque);if (addr >= q->lqspi_cached_addr && addr <= q->lqspi_cached_addr +LQSPI_CACHE_SIZE -4 ) {uint8_t *retp = &q->lqspi_buf[addr - q->lqspi_cached_addr]; - ret =cpu_to_le32 (*(uint32_t *)retp); -DB_PRINT_L (1 ,"addr: %08x, data: %08x\n" , (unsigned )addr, - (unsigned )ret); -return ret; - }else { -lqspi_load_cache (opaque, addr); -return lqspi_read (opaque, addr, size); + *value =cpu_to_le32 (*(uint32_t *)retp); +DB_PRINT_L (1 ,"addr: %08" HWADDR_PRIx", data: %08" PRIx64"\n" , + addr, *value); +return MEMTX_OK ; } + +lqspi_load_cache (opaque, addr); +return lqspi_read (opaque, addr, value, size, attrs); }static const MemoryRegionOps lqspi_ops = { - .read = lqspi_read, + .read_with_attrs = lqspi_read, .endianness =DEVICE_NATIVE_ENDIAN , .valid = { .min_access_size =1 ,Signature · 1 line
2.20.1
PATCH v4 2/3 hw/ssi/xilinx_spips: Avoid AXI writes to the LQSPI linear memory
In reply to Philippe Mathieu-DaudéLei Sun found while auditing the code that a CPU write would trigger a NULL pointer dereference.
From UG1085 datasheet [*] AXI writes in this region are ignored and generates an AXI Slave Error (SLVERR).
Fix by implementing the write_with_attrs() handler. Return MEMTX_ERROR when the region is accessed (this error maps to an AXI slave error).
[*] https://www.xilinx.com/support/documentation/user_guides/ug1085-zynq-ultrascale-trm.pdf
---
v4: Fix typos (Francisco)
--- hw/ssi/xilinx_spips.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+)
hw/ssi/xilinx_spips.c +16 -0
@@ -1220,8 +1220,24 @@ static MemTxResult lqspi_read(void *opaque, hwaddr addr, uint64_t *value,return lqspi_read (opaque, addr, value, size, attrs); } +static MemTxResult lqspi_write (void *opaque,hwaddr offset,uint64_t value, +unsigned size,MemTxAttrs attrs) +{ +/* +* From UG1085, Chapter 24 (Quad-SPI controllers): +* - Writes are ignored +* - AXI writes generate an external AXI slave error (SLVERR) +*/ +qemu_log_mask (LOG_GUEST_ERROR ,"%s Unexpected %u-bit access to 0x%" PRIx64 +" (value: 0x%" PRIx64"\n" , + __func__, size <<3 , offset, value); + +return MEMTX_ERROR ; +} +static const MemoryRegionOps lqspi_ops = { .read_with_attrs = lqspi_read, + .write_with_attrs = lqspi_write, .endianness =DEVICE_NATIVE_ENDIAN , .valid = { .min_access_size =1 ,
Signature · 1 line
2.20.1
RFC v4 3/3 hw/ssi/xilinx_spips: Avoid out-of-bound access to lqspi_buf[]
In reply to Philippe Mathieu-DaudéBoth lqspi_read() and lqspi_load_cache() expect a 32-bit aligned address.
Set MemoryRegionOps.impl values to force 32-bit accesses, this way we are sure we do not access the lqspi_buf[] array out of bound.
---
Late friday patch...
hw/ssi/xilinx_spips.c | 4 ++++ 1 file changed, 4 insertions(+)
hw/ssi/xilinx_spips.c +4 -0
@@ -1243,6 +1243,10 @@ static const MemoryRegionOps lqspi_ops = { .min_access_size = 1, .max_access_size = 4 } + .impl = { + .min_access_size = 4, + .max_access_size = 4, + } };static void xilinx_spips_realize (DeviceState *dev,Error **errp)
Signature · 1 line
2.20.1
Re: [Qemu-arm] [RFC PATCH-for-4.1 v4 3/3] hw/ssi/xilinx_spips: Avoid out-of-bound access to lqspi_buf[]
Hi Philippe,
Quoted 6 lines · Philippe Mathieu-Daudé
Both
lqspi_read()andlqspi_load_cache()expect a 32-bit aligned address.Set MemoryRegionOps.impl values to force 32-bit accesses, this way we are sure we do not access the lqspi_buf[] array out of bound.
The patch and correction follows chapter 24 aswell (UG1085) so I suggest dropping 'RFC' in the subject.
Quoted 13 lines
Signed-off-by Philippe Mathieu-Daudé <[email protected]>---
Late friday patch...hw/ssi/xilinx_spips.c | 4 ++++ 1 file changed, 4 insertions(+)hw/ssi/xilinx_spips.c +0
-0
hw/ssi/xilinx_spips.c static const MemoryRegionOps lqspi_ops = { · line 1245
Quoted 3 lines
hw/ssi/xilinx_spips.c +0
-0@@ -1243,6 +1243,10 @@ static const MemoryRegionOps lqspi_ops = { .min_access_size = 1, .max_access_size = 4 }
A comma is missing on the line above (else the patch doesn't compile).
Best regards,
Francisco
Quoted 10 lines
hw/ssi/xilinx_spips.c +4
-0+ .impl = { + .min_access_size = 4, + .max_access_size = 4, + } };static void xilinx_spips_realize (DeviceState *dev,Error **errp)Signature · 1 line
2.20.1
Re: [Qemu-arm] [RFC PATCH-for-4.1 v4 3/3] hw/ssi/xilinx_spips: Avoid out-of-bound access to lqspi_buf[]
Quoted 6 lines · Francisco Iglesias
Hi Philippe,
Quoted 3 lines · Philippe Mathieu-Daudé
Both
lqspi_read()andlqspi_load_cache()expect a 32-bit aligned address.
Quoted 6 lines
Quoted 3 lines
Set MemoryRegionOps.impl values to force 32-bit accesses, this way we are sure we do not access the lqspi_buf[] array out of bound.
The patch and correction follows chapter 24 aswell (UG1085) so I suggest dropping 'RFC' in the subject.
OK.
Quoted 8 lines
Quoted 8 lines
Signed-off-by Philippe Mathieu-Daudé <[email protected]>---
Late friday patch...hw/ssi/xilinx_spips.c | 4 ++++ 1 file changed, 4 insertions(+)
Quoted 10 lines
hw/ssi/xilinx_spips.c static const MemoryRegionOps lqspi_ops = { · line 1245
Quoted 8 lines
hw/ssi/xilinx_spips.c +0
-0@@ -1243,6 +1243,10 @@ static const MemoryRegionOps lqspi_ops = { .min_access_size = 1, .max_access_size = 4 }A comma is missing on the line above (else the patch doesn't compile).
Sorry, I first added this block before '.valid' and tested it, then moved it and thought the trailing comma was not useful and forgot to test for the previous '.valid' field :/
Quoted 14 lines
Best regards,
FranciscoQuoted 10 lines
hw/ssi/xilinx_spips.c +4
-0+ .impl = { + .min_access_size = 4, + .max_access_size = 4, + } };static void xilinx_spips_realize (DeviceState *dev,Error **errp)Signature · 1 line
2.20.1
Patchew URL: https://patchew.org/QEMU/[email protected]/
Patchew URL: https://patchew.org/QEMU/[email protected]/
Hi,
This series failed build test on s390x host. Please find the details below.
=== TEST SCRIPT BEGIN ===
#!/bin/bash
# Testing script will be invoked under the git checkout with # HEAD pointing to a commit that has the patches applied on top of "base"
# branch set -e
echo
echo "=== ENV ==="
env
echo
echo "=== PACKAGES ==="
rpm -qa
echo
echo "=== UNAME ==="
uname -a
CC=$HOME/bin/cc
INSTALL=$PWD/install
BUILD=$PWD/build
mkdir -p $BUILD $INSTALL
SRC=$PWD
cd $BUILD
$SRC/configure --cc=$CC --prefix=$INSTALL
make -j4
# XXX: we need reliable clean up
# make check -j4 V=1 make install === TEST SCRIPT END ===
CC hw/ssi/aspeed_smc.o
CC hw/ssi/stm32f2xx_spi.o
CC hw/ssi/mss-spi.o
/var/tmp/patchew-tester-tmp-ghjjv86b/src/hw/ssi/xilinx_spips.c:1246:5: error: expected ‘}’ before ‘.’ token
1246 | .impl = {
| ^/var/tmp/patchew-tester-tmp-ghjjv86b/src/hw/ssi/xilinx_spips.c:1238:42: note: to match this ‘{’
The full log is available at
http://patchew.org/logs/[email protected]/testing.s390x/?type=message.
---
Email generated automatically by Patchew [https://patchew.org/].
Please send your feedback to [email protected]