Re: [Qemu-devel] [PATCH] net: cadence_gem: fix compilation error when debug is on

["Philippe Mathieu-Daudé" <philmd@redhat.com>]

Hi Ramon,

On 6/9/19 12:08 PM, Ramon Fried wrote:
> defining CADENCE_GEM_ERR_DEBUG causes compilation
> errors, fix that.
> 
> Signed-off-by: Ramon Fried <rfried.dev@gmail.com>
> ---
>  hw/net/cadence_gem.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/hw/net/cadence_gem.c b/hw/net/cadence_gem.c
> index 7f63411430..5cc5a71524 100644
> --- a/hw/net/cadence_gem.c
> +++ b/hw/net/cadence_gem.c
> @@ -982,8 +982,8 @@ static ssize_t gem_receive(NetClientState *nc, const uint8_t *buf, size_t size)
>              return -1;
>          }
>  
> -        DB_PRINT("copy %d bytes to 0x%x\n", MIN(bytes_to_copy, rxbufsize),
> -                rx_desc_get_buffer(s->rx_desc[q]));
> +        DB_PRINT("copy %d bytes to 0x%lx\n", MIN(bytes_to_copy, rxbufsize),
> +                rx_desc_get_buffer(s, s->rx_desc[q]));

Your patch fails on 32-bit hosts:

./hw/net/cadence_gem.c:987:18: error: format '%lx' expects argument of
type 'long unsigned int', but argument 4 has type 'uint64_t {aka long
long unsigned int}' [-Werror=format=]
         DB_PRINT("copy %d bytes to 0x%lx\n", MIN(bytes_to_copy, rxbufsize),
                  ^
./hw/net/cadence_gem.c:39:24: note: in definition of macro 'DB_PRINT'
     fprintf(stderr, ## __VA_ARGS__); \
                        ^
./hw/net/cadence_gem.c: In function 'gem_transmit':
./hw/net/cadence_gem.c:1160:26: error: format '%lx' expects argument of
type 'long unsigned int', but argument 5 has type 'unsigned int'
[-Werror=format=]
                 DB_PRINT("TX descriptor @ 0x%x too large: size 0x%x
space " \
                          ^
./hw/net/cadence_gem.c:39:24: note: in definition of macro 'DB_PRINT'
     fprintf(stderr, ## __VA_ARGS__); \
                        ^
cc1: all warnings being treated as errors

QEMU provides "HWADDR_PRIx" format for addresses, see for example few
lines earlier:

    DB_PRINT("read descriptor 0x%" HWADDR_PRIx "\n", packet_desc_addr);


>  
>          /* Copy packet data to emulated DMA buffer */
>          address_space_write(&s->dma_as, rx_desc_get_buffer(s, s->rx_desc[q]) +
> @@ -1156,7 +1156,7 @@ static void gem_transmit(CadenceGEMState *s)
>              if (tx_desc_get_length(desc) > sizeof(tx_packet) -
>                                                 (p - tx_packet)) {
>                  DB_PRINT("TX descriptor @ 0x%x too large: size 0x%x space " \
> -                         "0x%x\n", (unsigned)packet_desc_addr,
> +                         "0x%lx\n", (unsigned)packet_desc_addr,

Here the correct format seems to be "%zd" (difference of sizeof).

>                           (unsigned)tx_desc_get_length(desc),
>                           sizeof(tx_packet) - (p - tx_packet));
>                  break;
> 

Nowadays QEMU prefers to move from the old DB_PRINT() macros to the
trace events API, see for example this commit:

https://git.qemu.org/?p=qemu.git;a=commitdiff;h=da1804d17a9ed7f060c072fbc4889db5fbc9c7d2;hp=a4f667b6714916683408b983cfe0a615a725775f

The first line you changed would be replaced by a trace event, while the
second could be replaced by a qemu_log_mask() call (it is an error
condition).

Also I suggest to include "QEMU Trivial <qemu-trivial@nongnu.org>" in
the list of recipients, so your patch might get reviewed/merged quicker.

Regards,

Phil.