[PATCH] hw/core: properly terminate loading .hex on EOF record

Alex Bennée posted 1 patch 5 years, 7 months ago
Patches applied successfully (tree, apply log)
git fetch https://github.com/patchew-project/qemu tags/patchew/20200401193849.14017-1-alex.bennee@linaro.org
Test docker-mingw@fedora passed
Test docker-quick@centos7 passed
Test checkpatch passed
Test FreeBSD passed
Test asan passed
hw/core/loader.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
[PATCH] hw/core: properly terminate loading .hex on EOF record
Posted by Alex Bennée 5 years, 7 months ago
The https://makecode.microbit.org/#editor generates slightly weird
.hex files which work fine on a real microbit but causes QEMU to
choke. The reason is extraneous data after the EOF record which causes
the loader to attempt to write a bigger file than it should to the
"rom". According to the HEX file spec an EOF really should be the last
thing we process so lets do that.

Reported-by: Ursula Bennée <alex.bennee@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
---
 hw/core/loader.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/hw/core/loader.c b/hw/core/loader.c
index eeef6da9a1b..8bbb1797a4c 100644
--- a/hw/core/loader.c
+++ b/hw/core/loader.c
@@ -1447,6 +1447,7 @@ typedef struct {
     uint32_t current_rom_index;
     uint32_t rom_start_address;
     AddressSpace *as;
+    bool complete;
 } HexParser;
 
 /* return size or -1 if error */
@@ -1484,6 +1485,7 @@ static int handle_record_type(HexParser *parser)
                                   parser->current_rom_index,
                                   parser->rom_start_address, parser->as);
         }
+        parser->complete = true;
         return parser->total_size;
     case EXT_SEG_ADDR_RECORD:
     case EXT_LINEAR_ADDR_RECORD:
@@ -1548,11 +1550,12 @@ static int parse_hex_blob(const char *filename, hwaddr *addr, uint8_t *hex_blob,
         .bin_buf = g_malloc(hex_blob_size),
         .start_addr = addr,
         .as = as,
+        .complete = false
     };
 
     rom_transaction_begin();
 
-    for (; hex_blob < end; ++hex_blob) {
+    for (; hex_blob < end && !parser.complete; ++hex_blob) {
         switch (*hex_blob) {
         case '\r':
         case '\n':
-- 
2.20.1


Re: [PATCH] hw/core: properly terminate loading .hex on EOF record
Posted by Stefan Hajnoczi 5 years, 7 months ago
On Wed, Apr 01, 2020 at 08:38:49PM +0100, Alex Bennée wrote:
> The https://makecode.microbit.org/#editor generates slightly weird
> .hex files which work fine on a real microbit but causes QEMU to
> choke. The reason is extraneous data after the EOF record which causes
> the loader to attempt to write a bigger file than it should to the
> "rom". According to the HEX file spec an EOF really should be the last
> thing we process so lets do that.
> 
> Reported-by: Ursula Bennée <alex.bennee@linaro.org>
> Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
> ---
>  hw/core/loader.c | 5 ++++-
>  1 file changed, 4 insertions(+), 1 deletion(-)

Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Re: [PATCH] hw/core: properly terminate loading .hex on EOF record
Posted by Richard Henderson 5 years, 7 months ago
On 4/1/20 12:38 PM, Alex Bennée wrote:
> The https://makecode.microbit.org/#editor generates slightly weird
> .hex files which work fine on a real microbit but causes QEMU to
> choke. The reason is extraneous data after the EOF record which causes
> the loader to attempt to write a bigger file than it should to the
> "rom". According to the HEX file spec an EOF really should be the last
> thing we process so lets do that.
> 
> Reported-by: Ursula Bennée <alex.bennee@linaro.org>
> Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
> ---
>  hw/core/loader.c | 5 ++++-
>  1 file changed, 4 insertions(+), 1 deletion(-)

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>

r~

Re: [PATCH] hw/core: properly terminate loading .hex on EOF record
Posted by Philippe Mathieu-Daudé 5 years, 7 months ago
On 4/1/20 9:38 PM, Alex Bennée wrote:
> The https://makecode.microbit.org/#editor generates slightly weird
> .hex files which work fine on a real microbit but causes QEMU to
> choke. The reason is extraneous data after the EOF record which causes
> the loader to attempt to write a bigger file than it should to the
> "rom". According to the HEX file spec an EOF really should be the last
> thing we process so lets do that.
> 
> Reported-by: Ursula Bennée <alex.bennee@linaro.org>
> Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
> ---
>   hw/core/loader.c | 5 ++++-
>   1 file changed, 4 insertions(+), 1 deletion(-)
> 
> diff --git a/hw/core/loader.c b/hw/core/loader.c
> index eeef6da9a1b..8bbb1797a4c 100644
> --- a/hw/core/loader.c
> +++ b/hw/core/loader.c
> @@ -1447,6 +1447,7 @@ typedef struct {
>       uint32_t current_rom_index;
>       uint32_t rom_start_address;
>       AddressSpace *as;
> +    bool complete;
>   } HexParser;
>   
>   /* return size or -1 if error */
> @@ -1484,6 +1485,7 @@ static int handle_record_type(HexParser *parser)
>                                     parser->current_rom_index,
>                                     parser->rom_start_address, parser->as);
>           }
> +        parser->complete = true;
>           return parser->total_size;
>       case EXT_SEG_ADDR_RECORD:
>       case EXT_LINEAR_ADDR_RECORD:
> @@ -1548,11 +1550,12 @@ static int parse_hex_blob(const char *filename, hwaddr *addr, uint8_t *hex_blob,
>           .bin_buf = g_malloc(hex_blob_size),
>           .start_addr = addr,
>           .as = as,
> +        .complete = false
>       };
>   
>       rom_transaction_begin();
>   
> -    for (; hex_blob < end; ++hex_blob) {
> +    for (; hex_blob < end && !parser.complete; ++hex_blob) {
>           switch (*hex_blob) {
>           case '\r':
>           case '\n':
> 

Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>