Skip to contentSkip to conversation

block: Fix integer promotion error in bdrv_getlength()

PATCH Applied Eric Blake 3 participants 3 messages Nov 5, 2020

lore
Share & tools
lore

Eric Blake

author
#0 lore raw

PATCH v4 1/1 block: Fix integer promotion error in bdrv_getlength()

Back in 2015, we attempted to fix error reporting for images that claimed to have more than INT64_MAX/512 sectors, but due to the type promotions caused by BDRV_SECTOR_SIZE being unsigned, this inadvertently forces all negative ret values to be slammed into -EFBIG rather than the original error. While we're at it, we can avoid the confusing ?: by spelling the logic more directly.

Fixes 4a9c9ea0d3
---
 block.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)
block.c +7 -2
index 56bacc9e9f13..2fd932154e12 100644
@@ -5091,8 +5091,13 @@ int64_t bdrv_getlength(BlockDriverState *bs)
 {
     int64_t ret = bdrv_nb_sectors(bs);
 
-    ret = ret > INT64_MAX / BDRV_SECTOR_SIZE ? -EFBIG : ret;
-    return ret < 0 ? ret : ret * BDRV_SECTOR_SIZE;
+    if (ret < 0) {
+        return ret;
+    }
+    if (ret > INT64_MAX / BDRV_SECTOR_SIZE) {
+        return -EFBIG;
+    }
+    return ret * BDRV_SECTOR_SIZE;
 }
 
 /* return 0 as number of sectors if no device present or error */
Signature · 1 line
2.28.0
Reviewed-by on v4 1/1
#1 lore raw
Quoted 10 lines · Eric Blake

Back in 2015, we attempted to fix error reporting for images that claimed to have more than INT64_MAX/512 sectors, but due to the type promotions caused by BDRV_SECTOR_SIZE being unsigned, this inadvertently forces all negative ret values to be slammed into -EFBIG rather than the original error. While we're at it, we can avoid the confusing ?: by spelling the logic more directly.

Fixes 4a9c9ea0d3

Berto

Applied
#2 lore raw
In reply to Eric Blake
Quoted 7 lines · Eric Blake

Back in 2015, we attempted to fix error reporting for images that claimed to have more than INT64_MAX/512 sectors, but due to the type promotions caused by BDRV_SECTOR_SIZE being unsigned, this inadvertently forces all negative ret values to be slammed into -EFBIG rather than the original error. While we're at it, we can avoid the confusing ?: by spelling the logic more directly.

Quoted 6 lines
Fixes 4a9c9ea0d3
---
  block.c | 9 +++++++--
  1 file changed, 7 insertions(+), 2 deletions(-)

Thanks, applied to my block branch (replacing the original patch):

https://git.xanclic.moe/XanClic/qemu/commits/branch/block

Max