From: Palmer Dabbelt <palmer@sifive.com>
To: qemu-riscv@nongnu.org
Cc: Palmer Dabbelt <palmer@sifive.com>, qemu-devel@nongnu.org
Subject: [Qemu-devel] [PATCH] RISC-V: Fix a memory leak when realizing a sifive_e
Date: Fri, 14 Jun 2019 05:08:30 -0700 [thread overview]
Message-ID: <20190614120830.21850-1-palmer@sifive.com> (raw)
Coverity pointed out a memory leak in riscv_sifive_e_soc_realize(),
where a pair of recently added MemoryRegion instances would not be freed
if there were errors elsewhere in the function. The fix here is to
simply not use dynamic allocation for these instances: there's always
one of each in SiFiveESoCState, so instead we just include them within
the struct.
Thanks to Peter for pointing out the bug and suggesting the fix!
Fixes: 30efbf330a45 ("SiFive RISC-V GPIO Device")
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
---
hw/riscv/sifive_e.c | 12 +++++-------
include/hw/riscv/sifive_e.h | 2 ++
2 files changed, 7 insertions(+), 7 deletions(-)
diff --git a/hw/riscv/sifive_e.c b/hw/riscv/sifive_e.c
index 80ac56fa7d5e..83375afcd1d6 100644
--- a/hw/riscv/sifive_e.c
+++ b/hw/riscv/sifive_e.c
@@ -158,17 +158,15 @@ static void riscv_sifive_e_soc_realize(DeviceState *dev, Error **errp)
SiFiveESoCState *s = RISCV_E_SOC(dev);
MemoryRegion *sys_mem = get_system_memory();
- MemoryRegion *xip_mem = g_new(MemoryRegion, 1);
- MemoryRegion *mask_rom = g_new(MemoryRegion, 1);
object_property_set_bool(OBJECT(&s->cpus), true, "realized",
&error_abort);
/* Mask ROM */
- memory_region_init_rom(mask_rom, NULL, "riscv.sifive.e.mrom",
+ memory_region_init_rom(&s->mask_rom, NULL, "riscv.sifive.e.mrom",
memmap[SIFIVE_E_MROM].size, &error_fatal);
memory_region_add_subregion(sys_mem,
- memmap[SIFIVE_E_MROM].base, mask_rom);
+ memmap[SIFIVE_E_MROM].base, &s->mask_rom);
/* MMIO */
s->plic = sifive_plic_create(memmap[SIFIVE_E_PLIC].base,
@@ -228,10 +226,10 @@ static void riscv_sifive_e_soc_realize(DeviceState *dev, Error **errp)
memmap[SIFIVE_E_PWM2].base, memmap[SIFIVE_E_PWM2].size);
/* Flash memory */
- memory_region_init_ram(xip_mem, NULL, "riscv.sifive.e.xip",
+ memory_region_init_ram(&s->xip_mem, NULL, "riscv.sifive.e.xip",
memmap[SIFIVE_E_XIP].size, &error_fatal);
- memory_region_set_readonly(xip_mem, true);
- memory_region_add_subregion(sys_mem, memmap[SIFIVE_E_XIP].base, xip_mem);
+ memory_region_set_readonly(&s->xip_mem, true);
+ memory_region_add_subregion(sys_mem, memmap[SIFIVE_E_XIP].base, &s->xip_mem);
}
static void riscv_sifive_e_machine_init(MachineClass *mc)
diff --git a/include/hw/riscv/sifive_e.h b/include/hw/riscv/sifive_e.h
index 3b14eb74621f..d175b24cb209 100644
--- a/include/hw/riscv/sifive_e.h
+++ b/include/hw/riscv/sifive_e.h
@@ -33,6 +33,8 @@ typedef struct SiFiveESoCState {
RISCVHartArrayState cpus;
DeviceState *plic;
SIFIVEGPIOState gpio;
+ MemoryRegion xip_mem;
+ MemoryRegion mask_rom;
} SiFiveESoCState;
typedef struct SiFiveEState {
--
2.21.0
next reply other threads:[~2019-06-14 12:20 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-06-14 12:08 Palmer Dabbelt [this message]
2019-06-14 12:25 ` [Qemu-devel] [PATCH] RISC-V: Fix a memory leak when realizing a sifive_e Philippe Mathieu-Daudé
2019-06-16 8:15 ` Palmer Dabbelt
2019-06-17 17:18 ` Alistair Francis
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20190614120830.21850-1-palmer@sifive.com \
--to=palmer@sifive.com \
--cc=qemu-devel@nongnu.org \
--cc=qemu-riscv@nongnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).