qemu/vvfat-Fix-vvfat_write-for-writes-before-the-root-dir.patch
Jiabo Feng f4f53a5098 QEMU update to version 6.2.0-106:
- hw/nvme: Remove redundant dma_blk_write
- tests/avocado/machine_s390_ccw_virtio: Adapt test to new default resolution
- edid: set default resolution to 1280x800 (WXGA)
- iotests/308: Fix for CAP_DAC_OVERRIDE
- hvf: remove unused but set variable
- vvfat: Fix vvfat_write() for writes before the root directory
- hw/misc/nrf51_rng: Don't use BIT_MASK() when we mean BIT()
- hw/pci: Remove unused pci_irq_pulse() method
- ui/gtk: fix leaks found wtih fuzzing
- target/i386: fix size of EBP writeback in gen_enter()
- tests/qtest/fuzz: fix memleak in qos_fuzz.c
- hw/core/loader: gunzip(): fix memory leak on error path
- migration: fix a typo
- scsi: fetch unit attention when creating the request
- raw-format: Fix error message for invalid offset/size
- tcg: Reset data_gen_ptr correctly
- Fix calculation of minimum in colo_compare_tcp
- hw/intc: Don't clear pending bits on IRQ lowering
- target/arm: Drop user-only special case in sve_stN_r
- usb-hub: Fix handling port power control messages
- target/ppc: Set ctx->opcode for decode_insn32()
- linux-user: Add proper strace format strings for getdents()/getdents64()
- linux-user: Fix TARGET_PROT_SEM for XTENSA
- linux-user/hppa: Set TASK_UNMAPPED_BASE to 0xfa000000 for hppa arch
- linux-user/hppa: Dump IIR on register dump
- tests: Fix typo in check-help output
- qdev-core.h: Fix wrongly named reference to TYPE_SPLIT_IRQ
- hw/scsi/megasas: Simplify using the ldst API
- gqa-win: get_pci_info: Clean dev_info if handle is valid
- target/ppc: Fix 7448 support
- vvfat: Fix size of temporary qcow file
- docs: Correct 'vhost-user-blk' spelling
- jackaudio: use ifdefs to hide unavailable functions
- simplebench: Fix Python syntax error (reported by LGTM)
- python: update type hints for mypy 0.930
- Python/aqmp: fix type definitions for mypy 0.920
- tcg: Allow top bit of SIMD_DATA_BITS to be set in simd_desc()
- hw/ppc/e500: Prefer QOM cast
- hw/ppc/e500: Remove unused "irqs" parameter
- hw/ppc/e500: Add missing device tree properties to i2c controller node
- linux-user: Show timespec on strace for futex()
- linux-user: Add strace for clock_nanosleep()
- linux-user: Fix strace of chmod() if mode == 0
- linux-user: Log failing executable in EXCP_DUMP()
- linux-user: Add pidfd_open(), pidfd_send_signal() and pidfd_getfd() syscalls

Signed-off-by: Jiabo Feng <fengjiabo1@huawei.com>
(cherry picked from commit 87ebac5b5cfb97ddb7ac2af097703758fb0751c4)
2025-02-21 17:43:48 +08:00

100 lines
3.3 KiB
Diff

From 64b53f59bccb8ec3251826c06d74adbc7b3cad36 Mon Sep 17 00:00:00 2001
From: Luo Yifan <luoyifan_yewu@cmss.chinamobile.com>
Date: Mon, 16 Dec 2024 10:27:44 +0800
Subject: [PATCH] vvfat: Fix vvfat_write() for writes before the root directory
cherry picked from commit b9b8860d24676ec59c878d5206ea6bcfc87af798
The calculation in sector2cluster() is done relative to the offset of
the root directory. Any writes to blocks before the start of the root
directory (in particular, writes to the FAT) result in negative values,
which are not handled correctly in vvfat_write().
This changes sector2cluster() to return a signed value, and makes sure
that vvfat_write() doesn't try to find mappings for negative cluster
number. It clarifies the code in vvfat_write() to make it more obvious
that the cluster numbers can be negative.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20211209152231.23756-1-kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Luo Yifan <luoyifan_yewu@cmss.chinamobile.com>
---
block/vvfat.c | 30 ++++++++++++++++++++++--------
1 file changed, 22 insertions(+), 8 deletions(-)
diff --git a/block/vvfat.c b/block/vvfat.c
index 3691c4774e..935a10bdd3 100644
--- a/block/vvfat.c
+++ b/block/vvfat.c
@@ -882,7 +882,7 @@ static int read_directory(BDRVVVFATState* s, int mapping_index)
return 0;
}
-static inline uint32_t sector2cluster(BDRVVVFATState* s,off_t sector_num)
+static inline int32_t sector2cluster(BDRVVVFATState* s,off_t sector_num)
{
return (sector_num - s->offset_to_root_dir) / s->sectors_per_cluster;
}
@@ -2983,6 +2983,7 @@ static int vvfat_write(BlockDriverState *bs, int64_t sector_num,
{
BDRVVVFATState *s = bs->opaque;
int i, ret;
+ int first_cluster, last_cluster;
DLOG(checkpoint());
@@ -3001,9 +3002,20 @@ DLOG(checkpoint());
if (sector_num < s->offset_to_fat)
return -1;
- for (i = sector2cluster(s, sector_num);
- i <= sector2cluster(s, sector_num + nb_sectors - 1);) {
- mapping_t* mapping = find_mapping_for_cluster(s, i);
+ /*
+ * Values will be negative for writes to the FAT, which is located before
+ * the root directory.
+ */
+ first_cluster = sector2cluster(s, sector_num);
+ last_cluster = sector2cluster(s, sector_num + nb_sectors - 1);
+
+ for (i = first_cluster; i <= last_cluster;) {
+ mapping_t *mapping = NULL;
+
+ if (i >= 0) {
+ mapping = find_mapping_for_cluster(s, i);
+ }
+
if (mapping) {
if (mapping->read_only) {
fprintf(stderr, "Tried to write to write-protected file %s\n",
@@ -3043,8 +3055,9 @@ DLOG(checkpoint());
}
}
i = mapping->end;
- } else
+ } else {
i++;
+ }
}
/*
@@ -3058,10 +3071,11 @@ DLOG(fprintf(stderr, "Write to qcow backend: %d + %d\n", (int)sector_num, nb_sec
return ret;
}
- for (i = sector2cluster(s, sector_num);
- i <= sector2cluster(s, sector_num + nb_sectors - 1); i++)
- if (i >= 0)
+ for (i = first_cluster; i <= last_cluster; i++) {
+ if (i >= 0) {
s->used_clusters[i] |= USED_ALLOCATED;
+ }
+ }
DLOG(checkpoint());
/* TODO: add timeout */
--
2.41.0.windows.1