qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Bernhard Beschow <shentey@gmail.com>
To: "BALATON Zoltan" <balaton@eik.bme.hu>,
	"Cédric Le Goater" <clg@redhat.com>
Cc: qemu-devel@nongnu.org, "Hanna Reitz" <hreitz@redhat.com>,
	qemu-ppc@nongnu.org, "Kevin Wolf" <kwolf@redhat.com>,
	"Corey Minyard" <cminyard@mvista.com>,
	"Philippe Mathieu-Daudé" <philmd@linaro.org>,
	"Paolo Bonzini" <pbonzini@redhat.com>,
	"Alex Williamson" <alex.williamson@redhat.com>,
	"Jason Wang" <jasowang@redhat.com>,
	"Daniel Henrique Barboza" <danielhb413@gmail.com>,
	qemu-block@nongnu.org, "Nicholas Piggin" <npiggin@gmail.com>,
	"Bin Meng" <bmeng.cn@gmail.com>
Subject: Re: [PATCH 01/23] hw/ppc/e500: Do not leak struct boot_info
Date: Tue, 01 Oct 2024 14:26:26 +0000	[thread overview]
Message-ID: <2F9AA8B2-BF07-4F9C-BA0E-6403BFADF75D@gmail.com> (raw)
In-Reply-To: <c823aedf-767f-6ddf-31f1-16ba667abeff@eik.bme.hu>



Am 26. September 2024 00:14:04 UTC schrieb BALATON Zoltan <balaton@eik.bme.hu>:
>On Wed, 25 Sep 2024, Cédric Le Goater wrote:
>> On 9/23/24 11:29, Bernhard Beschow wrote:
>>> The struct is allocated once with g_new0() but never free()'d. Fix the leakage
>>> by adding an attribute to struct PPCE500MachineState which avoids the
>>> allocation.
>>> 
>>> Signed-off-by: Bernhard Beschow <shentey@gmail.com>
>>> ---
>>>   hw/ppc/e500.h |  8 ++++++++
>>>   hw/ppc/e500.c | 17 ++++-------------
>>>   2 files changed, 12 insertions(+), 13 deletions(-)
>>> 
>>> diff --git a/hw/ppc/e500.h b/hw/ppc/e500.h
>>> index 8c09ef92e4..557ce6ad93 100644
>>> --- a/hw/ppc/e500.h
>>> +++ b/hw/ppc/e500.h
>>> @@ -5,10 +5,18 @@
>>>   #include "hw/platform-bus.h"
>>>   #include "qom/object.h"
>>>   +typedef struct boot_info {
>>> +    uint32_t dt_base;
>>> +    uint32_t dt_size;
>>> +    uint32_t entry;
>>> +} boot_info;
>> 
>> or simply move the fields under the machine state struct to avoif
>> the struct boot_info which doesn't seem that necessary. Is it ?
>
>It's passed to CPU reset function via env->load_info. It could be possible to pass the whole machine state but it seems that's unneeded so this struct just contains what's needed for this. Other machines also have similar boot_info structs although they seem to be different and not common to all machines. Thus I don't think merging with machine state would be better than keeping is separate as this is more CPU related.

I agree that having a consistent style across machines is prefereable, so I'll keep struct boot_info. It also avoids conflicts with your TLB cleanup series.

Best regards,
Bernhard

>
>Regards,
>BALATON Zoltan
>
>> 
>> Thanks,
>> 
>> C.
>> 
>> 
>> 
>>> +
>>>   struct PPCE500MachineState {
>>>       /*< private >*/
>>>       MachineState parent_obj;
>>>   +    boot_info boot_info;
>>> +
>>>       /* points to instance of TYPE_PLATFORM_BUS_DEVICE if
>>>        * board supports dynamic sysbus devices
>>>        */
>>> diff --git a/hw/ppc/e500.c b/hw/ppc/e500.c
>>> index 3bd12b54ab..75b051009f 100644
>>> --- a/hw/ppc/e500.c
>>> +++ b/hw/ppc/e500.c
>>> @@ -80,13 +80,6 @@
>>>     #define PLATFORM_CLK_FREQ_HZ       (400 * 1000 * 1000)
>>>   -struct boot_info
>>> -{
>>> -    uint32_t dt_base;
>>> -    uint32_t dt_size;
>>> -    uint32_t entry;
>>> -};
>>> -
>>>   static uint32_t *pci_map_create(void *fdt, uint32_t mpic, int first_slot,
>>>                                   int nr_slots, int *len)
>>>   {
>>> @@ -919,7 +912,6 @@ void ppce500_init(MachineState *machine)
>>>       bool kernel_as_payload;
>>>       hwaddr bios_entry = 0;
>>>       target_long payload_size;
>>> -    struct boot_info *boot_info = NULL;
>>>       int dt_size;
>>>       int i;
>>>       unsigned int smp_cpus = machine->smp.cpus;
>>> @@ -974,9 +966,8 @@ void ppce500_init(MachineState *machine)
>>>           /* Register reset handler */
>>>           if (!i) {
>>>               /* Primary CPU */
>>> -            boot_info = g_new0(struct boot_info, 1);
>>>               qemu_register_reset(ppce500_cpu_reset, cpu);
>>> -            env->load_info = boot_info;
>>> +            env->load_info = &pms->boot_info;
>>>           } else {
>>>               /* Secondary CPUs */
>>>               qemu_register_reset(ppce500_cpu_reset_sec, cpu);
>>> @@ -1274,9 +1265,9 @@ void ppce500_init(MachineState *machine)
>>>       }
>>>       assert(dt_size < DTB_MAX_SIZE);
>>>   -    boot_info->entry = bios_entry;
>>> -    boot_info->dt_base = dt_base;
>>> -    boot_info->dt_size = dt_size;
>>> +    pms->boot_info.entry = bios_entry;
>>> +    pms->boot_info.dt_base = dt_base;
>>> +    pms->boot_info.dt_size = dt_size;
>>>   }
>>>     static void e500_ccsr_initfn(Object *obj)
>> 
>> 
>>


  reply	other threads:[~2024-10-01 14:29 UTC|newest]

Thread overview: 70+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-09-23  9:29 [PATCH 00/23] E500 Cleanup Bernhard Beschow
2024-09-23  9:29 ` [PATCH 01/23] hw/ppc/e500: Do not leak struct boot_info Bernhard Beschow
2024-09-23 10:02   ` BALATON Zoltan
2024-09-25 19:08     ` Bernhard Beschow
2024-09-25 15:35   ` Cédric Le Goater
2024-09-25 19:03     ` Bernhard Beschow
2024-09-26  0:14     ` BALATON Zoltan
2024-10-01 14:26       ` Bernhard Beschow [this message]
2024-09-23  9:29 ` [PATCH 02/23] hw/ppc/e500: Reduce scope of env pointer Bernhard Beschow
2024-09-23 10:04   ` BALATON Zoltan
2024-09-25 19:09     ` Bernhard Beschow
2024-09-25 15:37   ` Cédric Le Goater
2024-09-25 19:02     ` Bernhard Beschow
2024-09-23  9:29 ` [PATCH 03/23] hw/ppc/e500: Prefer QOM cast Bernhard Beschow
2024-09-23 10:07   ` BALATON Zoltan
2024-09-23  9:29 ` [PATCH 04/23] hw/ppc/e500: Remove unused "irqs" parameter Bernhard Beschow
2024-09-23 10:18   ` BALATON Zoltan
2024-09-23  9:29 ` [PATCH 05/23] hw/ppc/e500: Add missing device tree properties to i2c controller node Bernhard Beschow
2024-09-25 15:37   ` Cédric Le Goater
2024-09-23  9:29 ` [PATCH 06/23] hw/ppc/e500: Use SysBusDevice API to access TYPE_CCSR's internal resources Bernhard Beschow
2024-09-23 10:28   ` BALATON Zoltan
2024-09-27 16:57     ` Bernhard Beschow
2024-09-23  9:30 ` [PATCH 07/23] hw/ppc/e500: Extract ppce500_ccsr.c Bernhard Beschow
2024-09-23 10:38   ` BALATON Zoltan
2024-09-24 20:02     ` Bernhard Beschow
2024-10-01 19:31       ` Bernhard Beschow
2024-09-23  9:30 ` [PATCH 08/23] hw/ppc/ppce500_ccsr: Log access to unimplemented registers Bernhard Beschow
2024-09-24 10:15   ` BALATON Zoltan
2024-09-24 19:23     ` Bernhard Beschow
2024-09-23  9:30 ` [PATCH 09/23] hw/ppc/mpc8544_guts: Populate POR PLL ratio status register Bernhard Beschow
2024-09-23 10:43   ` BALATON Zoltan
2024-09-23 21:54     ` Bernhard Beschow
2024-09-24  9:59       ` BALATON Zoltan
2024-09-24 19:13         ` Bernhard Beschow
2024-09-24 21:06           ` BALATON Zoltan
2024-09-23  9:30 ` [PATCH 10/23] hw/i2c/mpc_i2c: Convert DPRINTF to trace events for register access Bernhard Beschow
2024-09-25 15:40   ` Cédric Le Goater
2024-09-23  9:30 ` [PATCH 11/23] hw/i2c/mpc_i2c: Prefer DEFINE_TYPES() macro Bernhard Beschow
2024-09-23 10:49   ` BALATON Zoltan
2024-09-23 22:01     ` Bernhard Beschow
2024-09-24 10:12       ` BALATON Zoltan
2024-09-25 15:40   ` Cédric Le Goater
2024-09-23  9:30 ` [PATCH 12/23] hw/pci-host/ppce500: Reuse TYPE_PPC_E500_PCI_BRIDGE define Bernhard Beschow
2024-09-23 10:46   ` BALATON Zoltan
2024-09-23  9:30 ` [PATCH 13/23] hw/pci-host/ppce500: Prefer DEFINE_TYPES() macro Bernhard Beschow
2024-09-25 15:40   ` Cédric Le Goater
2024-09-23  9:30 ` [PATCH 14/23] hw/gpio/mpc8xxx: " Bernhard Beschow
2024-09-25 15:41   ` Cédric Le Goater
2024-09-23  9:30 ` [PATCH 15/23] hw/ppc/mpc8544_guts: " Bernhard Beschow
2024-09-25 15:41   ` Cédric Le Goater
2024-09-23  9:30 ` [PATCH 16/23] hw/net/fsl_etsec/etsec: " Bernhard Beschow
2024-09-25 15:41   ` Cédric Le Goater
2024-09-23  9:30 ` [PATCH 17/23] hw/intc: Guard openpic_kvm.c by dedicated OPENPIC_KVM Kconfig switch Bernhard Beschow
2024-09-25 15:41   ` Cédric Le Goater
2024-09-23  9:30 ` [PATCH 18/23] hw/sd/sdhci: Prefer DEFINE_TYPES() macro Bernhard Beschow
2024-09-25 15:41   ` Cédric Le Goater
2024-09-23  9:30 ` [PATCH 19/23] hw/block/pflash_cfi01: " Bernhard Beschow
2024-09-25 15:42   ` Cédric Le Goater
2024-09-23  9:30 ` [PATCH 20/23] hw/i2c/smbus_eeprom: " Bernhard Beschow
2024-09-25 15:42   ` Cédric Le Goater
2024-09-23  9:30 ` [PATCH 21/23] hw/rtc/ds1338: " Bernhard Beschow
2024-09-25 15:42   ` Cédric Le Goater
2024-09-23  9:30 ` [PATCH 22/23] hw/usb/hcd-ehci-sysbus: " Bernhard Beschow
2024-09-25 15:43   ` Cédric Le Goater
2024-09-23  9:30 ` [PATCH 23/23] hw/vfio/platform: Let vfio_start_eventfd_injection() take VFIOPlatformDevice pointer Bernhard Beschow
2024-09-24  8:36   ` Cédric Le Goater
2024-09-23 20:23 ` [PATCH 00/23] E500 Cleanup Cédric Le Goater
2024-09-23 21:25   ` Bernhard Beschow
2024-09-24  8:33     ` Cédric Le Goater
2024-09-26  9:15       ` Bernhard Beschow

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=2F9AA8B2-BF07-4F9C-BA0E-6403BFADF75D@gmail.com \
    --to=shentey@gmail.com \
    --cc=alex.williamson@redhat.com \
    --cc=balaton@eik.bme.hu \
    --cc=bmeng.cn@gmail.com \
    --cc=clg@redhat.com \
    --cc=cminyard@mvista.com \
    --cc=danielhb413@gmail.com \
    --cc=hreitz@redhat.com \
    --cc=jasowang@redhat.com \
    --cc=kwolf@redhat.com \
    --cc=npiggin@gmail.com \
    --cc=pbonzini@redhat.com \
    --cc=philmd@linaro.org \
    --cc=qemu-block@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-ppc@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).