Skip to contentSkip to conversation

Fix variadic formatting in blktap2 breaking xen-unstable building on x86_32

PATCH In review Tim Post 3 participants 4 messages Jun 26, 2009

lore
Share & tools
lore

Tim Post

author
#0 lore raw

PATCH 1/1 Fix variadic formatting in blktap2 breaking xen-unstable building on x86_32

Hi,

Building the tip of xen-unstable breaks, at least on x86_32 due to warnings being treated as errors with gcc 4.1

The problem is variadic formatting, uint64_t is unsigned long long, yet in several places in blktap2 it was formatted as unsigned long.

The attached patch fixes it.

I'm not sure where this was introduced, I just pulled the xen-unstable tip 15 minutes ago and noticed it.

Cheers,
--Tim

tools/blktap2/drivers/block-ram.c +2 -2
diff -r 442fb5f1e0a2 tools/blktap2/drivers/block-ram.c
@@ -101,7 +101,7 @@ static int get_image_info(int fd, td_dis
 	disksector_size = info->sector_size;
 	disksize        = info->size;
 	diskinfo        = info->info;
-	DPRINTF("Image sector_size: \n\t[%lu]\n",
+	DPRINTF("Image sector_size: \n\t[%llu]\n",
 		info->sector_size);
 
 	return 0;
@@ -126,7 +126,7 @@ int tdram_open (td_driver_t *driver, con
 			"sector_shift [%llu]\n",
 			(long long unsigned)(driver->info.size << SECTOR_SHIFT),
 			(long long unsigned)driver->info.size);
-		DPRINTF("Image sector_size: \n\t[%lu]\n",
+		DPRINTF("Image sector_size: \n\t[%llu]\n",
 			driver->info.sector_size);
 
 		prv->fd = -1;
tools/blktap2/drivers/block-vhd.c +1 -1
diff -r 442fb5f1e0a2 tools/blktap2/drivers/block-vhd.c
@@ -649,7 +649,7 @@ __vhd_open(td_driver_t *driver, const ch
 	driver->info.sector_size = VHD_SECTOR_SIZE;
 	driver->info.info        = 0;
 
-        DBG(TLOG_INFO, "vhd_open: done (sz:%"PRIu64", sct:%lu, inf:%u)\n",
+        DBG(TLOG_INFO, "vhd_open: done (sz:%"PRIu64", sct:%llu, inf:%u)\n",
 	    driver->info.size, driver->info.sector_size, driver->info.info);
 
 	if (test_vhd_flag(flags, VHD_FLAG_OPEN_STRICT) && 

_______________________________________________
Xen-devel mailing list
[email protected]
http://lists.xensource.com/xen-devel

#1 lore raw
Quoted 17 lines · Tim Post

Hi,

Building the tip of xen-unstable breaks, at least on x86_32 due to warnings being treated as errors with gcc 4.1

The problem is variadic formatting, uint64_t is unsigned long long, yet in several places in blktap2 it was formatted as unsigned long.

The attached patch fixes it.

I'm not sure where this was introduced, I just pulled the xen-unstable tip 15 minutes ago and noticed it.

diff -r 442fb5f1e0a2 tools/blktap2/drivers/block-ram.c
+++ b/tools/blktap2/drivers/block-ram.c Sat Jun 27 01:12:22 2009 +0800

tools/blktap2/drivers/block-ram.c static int get_image_info(int fd, td_dis · line 105

Quoted 6 lines
tools/blktap2/drivers/block-ram.c +1 -1
@@ -101,7 +101,7 @@ static int get_image_info(int fd, td_dis
 	disksector_size = info->sector_size;
 	disksize        = info->size;
 	diskinfo        = info->info;
-	DPRINTF("Image sector_size: \n\t[%lu]\n",
+	DPRINTF("Image sector_size: \n\t[%llu]\n",
 		info->sector_size);

This just ping-pongs between breaking 32-bit and 64-bit. I think you want to use PRIu64 in all these cases.

author
#2 lore raw
Quoted 2 lines · Brendan Cully

This just ping-pongs between breaking 32-bit and 64-bit. I think you want to use PRIu64 in all these cases.

Yes, sorry for sending without really looking, was leaving my office at the time. Several macros also need fixed, I'll re-send the patch tomorrow morning (its 2:39 AM my time now) .. unless someone else does it first.

Cheers,
--Tim

#3 lore raw

Hi Tim,

Quoted 9 lines · Tim Post
Quoted 2 lines · Brendan Cully

This just ping-pongs between breaking 32-bit and 64-bit. I think you want to use PRIu64 in all these cases.

Yes, sorry for sending without really looking, was leaving my office at the time. Several macros also need fixed, I'll re-send the patch tomorrow morning (its 2:39 AM my time now) .. unless someone else does it first.

I believe that this was filed by c/s 19826

cheers,
Ryan

Signature · 1 line
 Ryan O'Connor <[email protected]>